ruff fixes and restore default map location of object serializer load

- temporarily disable vram cache

Makefile

@@ -18,6 +18,7 @@ help:
 	@echo "frontend-typegen         Generate types for the frontend from the OpenAPI schema"
 	@echo "installer-zip            Build the installer .zip file for the current version"
 	@echo "tag-release              Tag the GitHub repository with the current version (use at release time only!)"
+	@echo "openapi                  Generate the OpenAPI schema for the app, outputting to stdout"
 
 # Runs ruff, fixing any safely-fixable errors and formatting
 ruff:
@@ -70,3 +71,6 @@ installer-zip:
 tag-release:
 	cd installer && ./tag_release.sh
 
+# Generate the OpenAPI Schema for the app
+openapi:
+	python scripts/generate_openapi_schema.py

docker/README.md

@@ -64,7 +64,7 @@ GPU_DRIVER=nvidia
 
 Any environment variables supported by InvokeAI can be set here - please see the [Configuration docs](https://invoke-ai.github.io/InvokeAI/features/CONFIGURATION/) for further detail.
 
-## Even Moar Customizing!
+## Even More Customizing!
 
 See the `docker-compose.yml` file. The `command` instruction can be uncommented and used to run arbitrary startup commands. Some examples below.
 

docs/contributing/DOWNLOAD_QUEUE.md

@@ -128,7 +128,8 @@ The queue operates on a series of download job objects. These objects
 specify the source and destination of the download, and keep track of
 the progress of the download.
 
-The only job type currently implemented is `DownloadJob`, a pydantic object with the
+Two job types are defined. `DownloadJob` and
+`MultiFileDownloadJob`. The former is a pydantic object with the
 following fields:
 
 | **Field**      | **Type**        |  **Default**  | **Description** |
@@ -138,7 +139,7 @@ following fields:
 | `dest`           | Path            |               | Where to download to              |
 | `access_token`   | str             |               | [optional] string containing authentication token for access |
 | `on_start`       | Callable        |               | [optional] callback when the download starts |
-| `on_progress`    | Callable        |               | [optional] callback called at intervals during download progress |
+| `on_progress` | Callable | | [optional] callback called at intervals during download progress |
 | `on_complete`    | Callable        |               | [optional] callback called after successful download completion |
 | `on_error`       | Callable        |               | [optional] callback called after an error occurs  |
 | `id`             | int             | auto assigned | Job ID, an integer >= 0           |
@@ -190,6 +191,33 @@ A cancelled job will have status `DownloadJobStatus.ERROR` and an
 `error_type` field of "DownloadJobCancelledException". In addition,
 the job's `cancelled` property will be set to True.
 
+The `MultiFileDownloadJob` is used for diffusers model downloads,
+which contain multiple files and directories under a common root:
+
+| **Field**      | **Type**        |  **Default**  | **Description** |
+|----------------|-----------------|---------------|-----------------|
+| _Fields passed in at job creation time_                               |
+| `download_parts` | Set[DownloadJob]|               | Component download jobs |
+| `dest`           | Path            |               | Where to download to              |
+| `on_start`       | Callable        |               | [optional] callback when the download starts |
+| `on_progress` | Callable | | [optional] callback called at intervals during download progress |
+| `on_complete`    | Callable        |               | [optional] callback called after successful download completion |
+| `on_error`       | Callable        |               | [optional] callback called after an error occurs  |
+| `id`             | int             | auto assigned | Job ID, an integer >= 0           |
+| _Fields updated over the course of the download task_
+| `status`         | DownloadJobStatus|              | Status code                                |
+| `download_path`  | Path |              | Path to the root of the downloaded files |
+| `bytes`          | int              | 0            | Bytes downloaded so far   |
+| `total_bytes`    | int              | 0            | Total size of the file at the remote site  |
+| `error_type`     | str              |              | String version of the exception that caused an error during download |
+| `error`          | str              |              | String version of the traceback associated with an error |
+| `cancelled`      | bool             | False        | Set to true if the job was cancelled by the caller|
+
+Note that the MultiFileDownloadJob does not support the `priority`,
+`job_started`, `job_ended` or `content_type` attributes. You can get
+these from the individual download jobs in `download_parts`.
+
+
 ### Callbacks
 
 Download jobs can be associated with a series of callbacks, each with
@@ -251,11 +279,40 @@ jobs using `list_jobs()`, fetch a single job by its with
 running jobs with `cancel_all_jobs()`, and wait for all jobs to finish
 with `join()`.
 
-#### job = queue.download(source, dest, priority, access_token)
+#### job = queue.download(source, dest, priority, access_token, on_start, on_progress, on_complete, on_cancelled, on_error)
 
 Create a new download job and put it on the queue, returning the
 DownloadJob object.
 
+#### multifile_job = queue.multifile_download(parts, dest, access_token, on_start, on_progress, on_complete, on_cancelled, on_error)
+
+This is similar to download(), but instead of taking a single source,
+it accepts a `parts` argument consisting of a list of
+`RemoteModelFile` objects. Each part corresponds to a URL/Path pair,
+where the URL is the location of the remote file, and the Path is the
+destination.
+
+`RemoteModelFile` can be imported from `invokeai.backend.model_manager.metadata`, and
+consists of a url/path pair. Note that the path *must* be relative.
+
+The method returns a `MultiFileDownloadJob`.
+
+
+```
+from invokeai.backend.model_manager.metadata import RemoteModelFile
+remote_file_1 = RemoteModelFile(url='http://www.foo.bar/my/pytorch_model.safetensors'',
+                                path='my_model/textencoder/pytorch_model.safetensors'
+			 			  )
+remote_file_2 = RemoteModelFile(url='http://www.bar.baz/vae.ckpt',
+                                path='my_model/vae/diffusers_model.safetensors'
+			 			  )
+job = queue.multifile_download(parts=[remote_file_1, remote_file_2],
+                               dest='/tmp/downloads',
+                               on_progress=TqdmProgress().update)
+queue.wait_for_job(job)
+print(f"The files were downloaded to {job.download_path}")
+```
+
 #### jobs = queue.list_jobs()
 
 Return a list of all active and inactive `DownloadJob`s.

docs/contributing/MODEL_MANAGER.md

@@ -397,26 +397,25 @@ In the event you wish to create a new installer, you may use the
 following initialization pattern:
 
 ```
-from invokeai.app.services.config import InvokeAIAppConfig
+from invokeai.app.services.config import get_config
 from invokeai.app.services.model_records import ModelRecordServiceSQL
 from invokeai.app.services.model_install import ModelInstallService
 from invokeai.app.services.download import DownloadQueueService
-from invokeai.app.services.shared.sqlite import SqliteDatabase
+from invokeai.app.services.shared.sqlite.sqlite_database import SqliteDatabase
 from invokeai.backend.util.logging import InvokeAILogger
 
-config = InvokeAIAppConfig.get_config()
-config.parse_args()
+config = get_config()
 
 logger = InvokeAILogger.get_logger(config=config)
-db = SqliteDatabase(config, logger)
+db = SqliteDatabase(config.db_path, logger)
 record_store = ModelRecordServiceSQL(db)
 queue = DownloadQueueService()
 queue.start()
 
-installer = ModelInstallService(app_config=config, 
+installer = ModelInstallService(app_config=config,
                                 record_store=record_store,
-              download_queue=queue
-           )
+                                download_queue=queue
+                                )
 installer.start()
 ```
 
@@ -1329,7 +1328,7 @@ from invokeai.app.services.model_load import ModelLoadService, ModelLoaderRegist
 
 config = InvokeAIAppConfig.get_config()
 ram_cache = ModelCache(
- max_cache_size=config.ram_cache_size, max_vram_cache_size=config.vram_cache_size, logger=logger
+ max_cache_size=config.ram_cache_size, logger=logger
 )
 convert_cache = ModelConvertCache(
  cache_path=config.models_convert_cache_path, max_size=config.convert_cache_size
@@ -1367,12 +1366,20 @@ the in-memory loaded model:
 | `model`        | AnyModel               | The instantiated model (details below) |
 | `locker`       | ModelLockerBase        | A context manager that mediates the movement of the model into VRAM |
 
-Because the loader can return multiple model types, it is typed to
-return `AnyModel`, a Union `ModelMixin`, `torch.nn.Module`,
-`IAIOnnxRuntimeModel`, `IPAdapter`, `IPAdapterPlus`, and
-`EmbeddingModelRaw`. `ModelMixin` is the base class of all diffusers
-models, `EmbeddingModelRaw` is used for LoRA and TextualInversion
-models. The others are obvious.
+### get_model_by_key(key, [submodel]) -> LoadedModel
+
+The `get_model_by_key()` method will retrieve the model using its
+unique database key. For example:
+
+loaded_model = loader.get_model_by_key('f13dd932c0c35c22dcb8d6cda4203764', SubModelType('vae'))
+
+`get_model_by_key()` may raise any of the following exceptions:
+
+* `UnknownModelException`   -- key not in database
+* `ModelNotFoundException`  -- key in database but model not found at path
+* `NotImplementedException` -- the loader doesn't know how to load this type of model
+
+### Using the Loaded Model in Inference
 
 `LoadedModel` acts as a context manager. The context loads the model
 into the execution device (e.g. VRAM on CUDA systems), locks the model
@@ -1380,17 +1387,33 @@ in the execution device for the duration of the context, and returns
 the model. Use it like this:
 
 ```
-model_info = loader.get_model_by_key('f13dd932c0c35c22dcb8d6cda4203764', SubModelType('vae'))
-with model_info as vae:
+loaded_model_= loader.get_model_by_key('f13dd932c0c35c22dcb8d6cda4203764', SubModelType('vae'))
+with loaded_model as vae:
  image = vae.decode(latents)[0]
 ```
 
-`get_model_by_key()` may raise any of the following exceptions:
+The object returned by the LoadedModel context manager is an
+`AnyModel`, which is a Union of `ModelMixin`, `torch.nn.Module`,
+`IAIOnnxRuntimeModel`, `IPAdapter`, `IPAdapterPlus`, and
+`EmbeddingModelRaw`. `ModelMixin` is the base class of all diffusers
+models, `EmbeddingModelRaw` is used for LoRA and TextualInversion
+models. The others are obvious.
+
+In addition, you may call `LoadedModel.model_on_device()`, a context
+manager that returns a tuple of the model's state dict in CPU and the
+model itself in VRAM. It is used to optimize the LoRA patching and
+unpatching process:
+
+```
+loaded_model_= loader.get_model_by_key('f13dd932c0c35c22dcb8d6cda4203764', SubModelType('vae'))
+with loaded_model.model_on_device() as (state_dict, vae):
+ image = vae.decode(latents)[0]
+```
+
+Since not all models have state dicts, the `state_dict` return value
+can be None.
+
 
-* `UnknownModelException`   -- key not in database
-* `ModelNotFoundException`  -- key in database but model not found at path
-* `NotImplementedException` -- the loader doesn't know how to load this type of model
-  
 ### Emitting model loading events
 
 When the `context` argument is passed to `load_model_*()`, it will
@@ -1578,3 +1601,59 @@ This method takes a model key, looks it up using the
 `ModelRecordServiceBase` object in `mm.store`, and passes the returned
 model configuration to `load_model_by_config()`.  It may raise a
 `NotImplementedException`.
+
+## Invocation Context Model Manager API
+
+Within invocations, the following methods are available from the
+`InvocationContext` object:
+
+### context.download_and_cache_model(source) -> Path
+
+This method accepts a `source` of a remote model, downloads and caches
+it locally, and then returns a Path to the local model. The source can
+be a direct download URL or a HuggingFace repo_id.
+
+In the case of HuggingFace repo_id, the following variants are
+recognized:
+
+* stabilityai/stable-diffusion-v4           -- default model
+* stabilityai/stable-diffusion-v4:fp16      -- fp16 variant
+* stabilityai/stable-diffusion-v4:fp16:vae  -- the fp16 vae subfolder
+* stabilityai/stable-diffusion-v4:onnx:vae  -- the onnx variant vae subfolder
+
+You can also point at an arbitrary individual file within a repo_id
+directory using this syntax:
+
+* stabilityai/stable-diffusion-v4::/checkpoints/sd4.safetensors
+
+### context.load_local_model(model_path, [loader]) -> LoadedModel
+
+This method loads a local model from the indicated path, returning a
+`LoadedModel`. The optional loader is a Callable that accepts a Path
+to the object, and returns a `AnyModel` object. If no loader is
+provided, then the method will use `torch.load()` for a .ckpt or .bin
+checkpoint file, `safetensors.torch.load_file()` for a safetensors
+checkpoint file, or `cls.from_pretrained()` for a directory that looks
+like a diffusers directory.
+
+### context.load_remote_model(source, [loader]) -> LoadedModel
+
+This method accepts a `source` of a remote model, downloads and caches
+it locally, loads it, and returns a `LoadedModel`. The source can be a
+direct download URL or a HuggingFace repo_id.
+
+In the case of HuggingFace repo_id, the following variants are
+recognized:
+
+* stabilityai/stable-diffusion-v4           -- default model
+* stabilityai/stable-diffusion-v4:fp16      -- fp16 variant
+* stabilityai/stable-diffusion-v4:fp16:vae  -- the fp16 vae subfolder
+* stabilityai/stable-diffusion-v4:onnx:vae  -- the onnx variant vae subfolder
+
+You can also point at an arbitrary individual file within a repo_id
+directory using this syntax:
+
+* stabilityai/stable-diffusion-v4::/checkpoints/sd4.safetensors
+
+
+

docs/contributing/frontend/WORKFLOWS.md

@@ -117,13 +117,13 @@ Stateless fields do not store their value in the node, so their field instances
 
 "Custom" fields will always be treated as stateless fields.
 
-##### Collection and Scalar Fields
+##### Single and Collection Fields
 
-Field types have a name and two flags which may identify it as a **collection** or **collection or scalar** field.
+Field types have a name and cardinality property which may identify it as a **SINGLE**, **COLLECTION** or **SINGLE_OR_COLLECTION** field.
 
-If a field is annotated in python as a list, its field type is parsed and flagged as a **collection** type (e.g. `list[int]`).
-
-If it is annotated as a union of a type and list, the type will be flagged as a **collection or scalar** type (e.g. `Union[int, list[int]]`). Fields may not be unions of different types (e.g. `Union[int, list[str]]` and `Union[int, str]` are not allowed).
+- If a field is annotated in python as a singular value or class, its field type is parsed as a **SINGLE** type (e.g. `int`, `ImageField`, `str`).
+- If a field is annotated in python as a list, its field type is parsed as a **COLLECTION** type (e.g. `list[int]`).
+- If it is annotated as a union of a type and list, the type will be parsed as a **SINGLE_OR_COLLECTION** type (e.g. `Union[int, list[int]]`). Fields may not be unions of different types (e.g. `Union[int, list[str]]` and `Union[int, str]` are not allowed).
 
 ## Implementation
 
@@ -173,8 +173,7 @@ Field types are represented as structured objects:
 ```ts
 type FieldType = {
   name: string;
-  isCollection: boolean;
-  isCollectionOrScalar: boolean;
+  cardinality: 'SINGLE' | 'COLLECTION' | 'SINGLE_OR_COLLECTION';
 };
 ```
 
@@ -186,7 +185,7 @@ There are 4 general cases for field type parsing.
 
 When a field is annotated as a primitive values (e.g. `int`, `str`, `float`), the field type parsing is fairly straightforward. The field is represented by a simple OpenAPI **schema object**, which has a `type` property.
 
-We create a field type name from this `type` string (e.g. `string` -> `StringField`).
+We create a field type name from this `type` string (e.g. `string` -> `StringField`). The cardinality is `"SINGLE"`.
 
 ##### Complex Types
 
@@ -200,13 +199,13 @@ We need to **dereference** the schema to pull these out. Dereferencing may requi
 
 When a field is annotated as a list of a single type, the schema object has an `items` property. They may be a schema object or reference object and must be parsed to determine the item type.
 
-We use the item type for field type name, adding `isCollection: true` to the field type.
+We use the item type for field type name. The cardinality is `"COLLECTION"`.
 
-##### Collection or Scalar Types
+##### Single or Collection Types
 
 When a field is annotated as a union of a type and list of that type, the schema object has an `anyOf` property, which holds a list of valid types for the union.
 
-After verifying that the union has two members (a type and list of the same type), we use the type for field type name, adding `isCollectionOrScalar: true` to the field type.
+After verifying that the union has two members (a type and list of the same type), we use the type for field type name, with cardinality `"SINGLE_OR_COLLECTION"`.
 
 ##### Optional Fields
 

docs/features/CONTROLNET.md

@@ -165,7 +165,7 @@ Additionally, each  section can be expanded with the "Show Advanced" button in o
 There are several ways to install IP-Adapter models with an existing InvokeAI installation:
 
 1. Through the command line interface launched from the invoke.sh / invoke.bat scripts, option [4] to download models.
-2. Through the Model Manager UI with models from the *Tools* section of [www.models.invoke.ai](https://www.models.invoke.ai). To do this, copy the repo ID from the desired model page, and paste it in the Add Model field of the model manager. **Note** Both the IP-Adapter and the Image Encoder must be installed for IP-Adapter to work. For example, the [SD 1.5 IP-Adapter](https://models.invoke.ai/InvokeAI/ip_adapter_plus_sd15) and [SD1.5 Image Encoder](https://models.invoke.ai/InvokeAI/ip_adapter_sd_image_encoder) must be installed to use IP-Adapter with SD1.5 based models.  
+2. Through the Model Manager UI with models from the *Tools* section of [models.invoke.ai](https://models.invoke.ai). To do this, copy the repo ID from the desired model page, and paste it in the Add Model field of the model manager. **Note** Both the IP-Adapter and the Image Encoder must be installed for IP-Adapter to work. For example, the [SD 1.5 IP-Adapter](https://models.invoke.ai/InvokeAI/ip_adapter_plus_sd15) and [SD1.5 Image Encoder](https://models.invoke.ai/InvokeAI/ip_adapter_sd_image_encoder) must be installed to use IP-Adapter with SD1.5 based models.  
 3. **Advanced -- Not recommended ** Manually downloading the IP-Adapter and Image Encoder files - Image Encoder folders shouid be placed in the `models\any\clip_vision` folders. IP Adapter Model folders should be placed in the relevant `ip-adapter` folder of relevant base model folder of Invoke root directory. For example, for the SDXL IP-Adapter, files should be added to the `model/sdxl/ip_adapter/` folder. 
 
 #### Using IP-Adapter

docs/help/FAQ.md

@@ -154,6 +154,18 @@ This is caused by an invalid setting in the `invokeai.yaml` configuration file.
 
 Check the [configuration docs] for more detail about the settings and how to specify them.
 
+## `ModuleNotFoundError: No module named 'controlnet_aux'`
+
+`controlnet_aux` is a dependency of Invoke and appears to have been packaged or distributed strangely. Sometimes, it doesn't install correctly. This is outside our control.
+
+If you encounter this error, the solution is to remove the package from the `pip` cache and re-run the Invoke installer so a fresh, working version of `controlnet_aux` can be downloaded and installed:
+
+- Run the Invoke launcher
+- Choose the developer console option
+- Run this command: `pip cache remove controlnet_aux`
+- Close the terminal window
+- Download and run the [installer](https://github.com/invoke-ai/InvokeAI/releases/latest), selecting your current install location
+
 ## Out of Memory Issues
 
 The models are large, VRAM is expensive, and you may find yourself

docs/help/diffusion.md

@@ -20,7 +20,7 @@ When you generate an image using text-to-image, multiple steps occur in latent s
 4. The VAE decodes the final latent image from latent space into image space.
 
 Image-to-image is a similar process, with only step 1 being different:
-1. The input image is encoded from image space into latent space by the VAE. Noise is then added to the input latent image. Denoising Strength dictates how may noise steps are added, and the amount of noise added at each step. A Denoising Strength of 0 means there are 0 steps and no noise added, resulting in an unchanged image, while a Denoising Strength of 1 results in the image being completely replaced with noise and a full set of denoising steps are performance. The process is then the same as steps 2-4 in the text-to-image process. 
+1. The input image is encoded from image space into latent space by the VAE. Noise is then added to the input latent image. Denoising Strength dictates how many noise steps are added, and the amount of noise added at each step. A Denoising Strength of 0 means there are 0 steps and no noise added, resulting in an unchanged image, while a Denoising Strength of 1 results in the image being completely replaced with noise and a full set of denoising steps are performance. The process is then the same as steps 2-4 in the text-to-image process. 
 
 Furthermore, a model provides the CLIP prompt tokenizer, the VAE, and a U-Net (where noise prediction occurs given a prompt and initial noise tensor).
 

docs/installation/010_INSTALL_AUTOMATED.md

@@ -98,7 +98,7 @@ Updating is exactly the same as installing - download the latest installer, choo
 
 If you have installation issues, please review the [FAQ]. You can also [create an issue] or ask for help on [discord].
 
-[installation requirements]: INSTALLATION.md#installation-requirements
+[installation requirements]: INSTALL_REQUIREMENTS.md
 [FAQ]: ../help/FAQ.md
 [install some models]: 050_INSTALLING_MODELS.md
 [configuration docs]: ../features/CONFIGURATION.md

docs/installation/020_INSTALL_MANUAL.md

@@ -10,7 +10,7 @@ InvokeAI is distributed as a python package on PyPI, installable with `pip`. The
 
 ### Requirements
 
-Before you start, go through the [installation requirements].
+Before you start, go through the [installation requirements](./INSTALL_REQUIREMENTS.md).
 
 ### Installation Walkthrough
 
@@ -79,7 +79,7 @@ Before you start, go through the [installation requirements].
 
 1. Install the InvokeAI Package. The base command is `pip install InvokeAI --use-pep517`, but you may need to change this depending on your system and the desired features.
 
-    - You may need to provide an [extra index URL]. Select your platform configuration using [this tool on the PyTorch website]. Copy the `--extra-index-url` string from this and append it to your install command.
+    - You may need to provide an [extra index URL](https://pip.pypa.io/en/stable/cli/pip_install/#cmdoption-extra-index-url). Select your platform configuration using [this tool on the PyTorch website](https://pytorch.org/get-started/locally/). Copy the `--extra-index-url` string from this and append it to your install command.
 
         !!! example "Install with an extra index URL"
 
@@ -116,4 +116,4 @@ Before you start, go through the [installation requirements].
 
     !!! warning
 
-        If the virtual environment is _not_ inside the root directory, then you _must_ specify the path to the root directory with `--root_dir \path\to\invokeai` or the `INVOKEAI_ROOT` environment variable.
+        If the virtual environment is _not_ inside the root directory, then you _must_ specify the path to the root directory with `--root \path\to\invokeai` or the `INVOKEAI_ROOT` environment variable.

docs/installation/INSTALL_REQUIREMENTS.md

@@ -37,13 +37,13 @@ Invoke runs best with a dedicated GPU, but will fall back to running on CPU, alb
     === "Nvidia"
 
         ```
-        Any GPU with at least 8GB VRAM. Linux only.
+        Any GPU with at least 8GB VRAM.
         ```
 
     === "AMD"
 
         ```
-        Any GPU with at least 16GB VRAM.
+        Any GPU with at least 16GB VRAM. Linux only.
         ```
 
     === "Mac"

installer/templates/invoke.bat.in

@@ -10,8 +10,7 @@ set INVOKEAI_ROOT=.
 echo Desired action:
 echo 1. Generate images with the browser-based interface
 echo 2. Open the developer console
-echo 3. Run the InvokeAI image database maintenance script
-echo 4. Command-line help
+echo 3. Command-line help
 echo Q - Quit
 echo.
 echo To update, download and run the installer from https://github.com/invoke-ai/InvokeAI/releases/latest.
@@ -34,9 +33,6 @@ IF /I "%choice%" == "1" (
     echo *** Type `exit` to quit this shell and deactivate the Python virtual environment ***
     call cmd /k
 ) ELSE IF /I "%choice%" == "3" (
-   echo Running the db maintenance script...
-   python .venv\Scripts\invokeai-db-maintenance.exe
-) ELSE IF /I "%choice%" == "4" (
     echo Displaying command line help...
     python .venv\Scripts\invokeai-web.exe --help %*
     pause

installer/templates/invoke.sh.in

@@ -47,11 +47,6 @@ do_choice() {
         bash --init-file "$file_name"
         ;;
     3)
-        clear
-        printf "Running the db maintenance script\n"
-        invokeai-db-maintenance --root ${INVOKEAI_ROOT}
-        ;;
-    4)
         clear
         printf "Command-line help\n"
         invokeai-web --help
@@ -71,8 +66,7 @@ do_line_input() {
     printf "What would you like to do?\n"
     printf "1: Generate images using the browser-based interface\n"
     printf "2: Open the developer console\n"
-    printf "3: Run the InvokeAI image database maintenance script\n"
-    printf "4: Command-line help\n"
+    printf "3: Command-line help\n"
     printf "Q: Quit\n\n"
     printf "To update, download and run the installer from https://github.com/invoke-ai/InvokeAI/releases/latest.\n\n"
     read -p "Please enter 1-4, Q: [1] " yn

invokeai/app/api/dependencies.py

@@ -18,6 +18,7 @@ from ..services.boards.boards_default import BoardService
 from ..services.bulk_download.bulk_download_default import BulkDownloadService
 from ..services.config import InvokeAIAppConfig
 from ..services.download import DownloadQueueService
+from ..services.events.events_fastapievents import FastAPIEventService
 from ..services.image_files.image_files_disk import DiskImageFileStorage
 from ..services.image_records.image_records_sqlite import SqliteImageRecordStorage
 from ..services.images.images_default import ImageService
@@ -29,11 +30,10 @@ from ..services.model_images.model_images_default import ModelImageFileStorageDi
 from ..services.model_manager.model_manager_default import ModelManagerService
 from ..services.model_records import ModelRecordServiceSQL
 from ..services.names.names_default import SimpleNameService
-from ..services.session_processor.session_processor_default import DefaultSessionProcessor
+from ..services.session_processor.session_processor_default import DefaultSessionProcessor, DefaultSessionRunner
 from ..services.session_queue.session_queue_sqlite import SqliteSessionQueue
 from ..services.urls.urls_default import LocalUrlService
 from ..services.workflow_records.workflow_records_sqlite import SqliteWorkflowRecordsStorage
-from .events import FastAPIEventService
 
 
 # TODO: is there a better way to achieve this?
@@ -93,7 +93,7 @@ class ApiDependencies:
         conditioning = ObjectSerializerForwardCache(
             ObjectSerializerDisk[ConditioningFieldData](output_folder / "conditioning", ephemeral=True)
         )
-        download_queue_service = DownloadQueueService(event_bus=events)
+        download_queue_service = DownloadQueueService(app_config=configuration, event_bus=events)
         model_images_service = ModelImageFileStorageDisk(model_images_folder / "model_images")
         model_manager = ModelManagerService.build_model_manager(
             app_config=configuration,
@@ -103,7 +103,7 @@ class ApiDependencies:
         )
         names = SimpleNameService()
         performance_statistics = InvocationStatsService()
-        session_processor = DefaultSessionProcessor()
+        session_processor = DefaultSessionProcessor(session_runner=DefaultSessionRunner())
         session_queue = SqliteSessionQueue(db=db)
         urls = LocalUrlService()
         workflow_records = SqliteWorkflowRecordsStorage(db=db)

invokeai/app/api/events.py

@@ -1,52 +0,0 @@
-# Copyright (c) 2022 Kyle Schouviller (https://github.com/kyle0654)
-
-import asyncio
-import threading
-from queue import Empty, Queue
-from typing import Any
-
-from fastapi_events.dispatcher import dispatch
-
-from ..services.events.events_base import EventServiceBase
-
-
-class FastAPIEventService(EventServiceBase):
-    event_handler_id: int
-    __queue: Queue
-    __stop_event: threading.Event
-
-    def __init__(self, event_handler_id: int) -> None:
-        self.event_handler_id = event_handler_id
-        self.__queue = Queue()
-        self.__stop_event = threading.Event()
-        asyncio.create_task(self.__dispatch_from_queue(stop_event=self.__stop_event))
-
-        super().__init__()
-
-    def stop(self, *args, **kwargs):
-        self.__stop_event.set()
-        self.__queue.put(None)
-
-    def dispatch(self, event_name: str, payload: Any) -> None:
-        self.__queue.put({"event_name": event_name, "payload": payload})
-
-    async def __dispatch_from_queue(self, stop_event: threading.Event):
-        """Get events on from the queue and dispatch them, from the correct thread"""
-        while not stop_event.is_set():
-            try:
-                event = self.__queue.get(block=False)
-                if not event:  # Probably stopping
-                    continue
-
-                dispatch(
-                    event.get("event_name"),
-                    payload=event.get("payload"),
-                    middleware_id=self.event_handler_id,
-                )
-
-            except Empty:
-                await asyncio.sleep(0.1)
-                pass
-
-            except asyncio.CancelledError as e:
-                raise e  # Raise a proper error

invokeai/app/api/routers/app_info.py

@@ -13,7 +13,6 @@ from pydantic import BaseModel, Field
 from invokeai.app.invocations.upscale import ESRGAN_MODELS
 from invokeai.app.services.invocation_cache.invocation_cache_common import InvocationCacheStatus
 from invokeai.backend.image_util.infill_methods.patchmatch import PatchMatch
-from invokeai.backend.image_util.safety_checker import SafetyChecker
 from invokeai.backend.util.logging import logging
 from invokeai.version import __version__
 
@@ -109,9 +108,7 @@ async def get_config() -> AppConfig:
         upscaling_models.append(str(Path(model).stem))
     upscaler = Upscaler(upscaling_method="esrgan", upscaling_models=upscaling_models)
 
-    nsfw_methods = []
-    if SafetyChecker.safety_checker_available():
-        nsfw_methods.append("nsfw_checker")
+    nsfw_methods = ["nsfw_checker"]
 
     watermarking_methods = ["invisible_watermark"]
 

invokeai/app/api/routers/images.py

@@ -6,13 +6,12 @@ from fastapi import BackgroundTasks, Body, HTTPException, Path, Query, Request,
 from fastapi.responses import FileResponse
 from fastapi.routing import APIRouter
 from PIL import Image
-from pydantic import BaseModel, Field, ValidationError
+from pydantic import BaseModel, Field, JsonValue
 
-from invokeai.app.invocations.fields import MetadataField, MetadataFieldValidator
+from invokeai.app.invocations.fields import MetadataField
 from invokeai.app.services.image_records.image_records_common import ImageCategory, ImageRecordChanges, ResourceOrigin
 from invokeai.app.services.images.images_common import ImageDTO, ImageUrlsDTO
 from invokeai.app.services.shared.pagination import OffsetPaginatedResults
-from invokeai.app.services.workflow_records.workflow_records_common import WorkflowWithoutID, WorkflowWithoutIDValidator
 
 from ..dependencies import ApiDependencies
 
@@ -42,13 +41,17 @@ async def upload_image(
     board_id: Optional[str] = Query(default=None, description="The board to add this image to, if any"),
     session_id: Optional[str] = Query(default=None, description="The session ID associated with this upload, if any"),
     crop_visible: Optional[bool] = Query(default=False, description="Whether to crop the image"),
+    metadata: Optional[JsonValue] = Body(
+        default=None, description="The metadata to associate with the image", embed=True
+    ),
 ) -> ImageDTO:
     """Uploads an image"""
     if not file.content_type or not file.content_type.startswith("image"):
         raise HTTPException(status_code=415, detail="Not an image")
 
-    metadata = None
-    workflow = None
+    _metadata = None
+    _workflow = None
+    _graph = None
 
     contents = await file.read()
     try:
@@ -62,22 +65,28 @@ async def upload_image(
 
     # TODO: retain non-invokeai metadata on upload?
     # attempt to parse metadata from image
-    metadata_raw = pil_image.info.get("invokeai_metadata", None)
-    if metadata_raw:
-        try:
-            metadata = MetadataFieldValidator.validate_json(metadata_raw)
-        except ValidationError:
-            ApiDependencies.invoker.services.logger.warn("Failed to parse metadata for uploaded image")
-            pass
+    metadata_raw = metadata if isinstance(metadata, str) else pil_image.info.get("invokeai_metadata", None)
+    if isinstance(metadata_raw, str):
+        _metadata = metadata_raw
+    else:
+        ApiDependencies.invoker.services.logger.debug("Failed to parse metadata for uploaded image")
+        pass
 
     # attempt to parse workflow from image
     workflow_raw = pil_image.info.get("invokeai_workflow", None)
-    if workflow_raw is not None:
-        try:
-            workflow = WorkflowWithoutIDValidator.validate_json(workflow_raw)
-        except ValidationError:
-            ApiDependencies.invoker.services.logger.warn("Failed to parse metadata for uploaded image")
-            pass
+    if isinstance(workflow_raw, str):
+        _workflow = workflow_raw
+    else:
+        ApiDependencies.invoker.services.logger.debug("Failed to parse workflow for uploaded image")
+        pass
+
+    # attempt to extract graph from image
+    graph_raw = pil_image.info.get("invokeai_graph", None)
+    if isinstance(graph_raw, str):
+        _graph = graph_raw
+    else:
+        ApiDependencies.invoker.services.logger.debug("Failed to parse graph for uploaded image")
+        pass
 
     try:
         image_dto = ApiDependencies.invoker.services.images.create(
@@ -86,8 +95,9 @@ async def upload_image(
             image_category=image_category,
             session_id=session_id,
             board_id=board_id,
-            metadata=metadata,
-            workflow=workflow,
+            metadata=_metadata,
+            workflow=_workflow,
+            graph=_graph,
             is_intermediate=is_intermediate,
         )
 
@@ -185,14 +195,21 @@ async def get_image_metadata(
         raise HTTPException(status_code=404)
 
 
+class WorkflowAndGraphResponse(BaseModel):
+    workflow: Optional[str] = Field(description="The workflow used to generate the image, as stringified JSON")
+    graph: Optional[str] = Field(description="The graph used to generate the image, as stringified JSON")
+
+
 @images_router.get(
-    "/i/{image_name}/workflow", operation_id="get_image_workflow", response_model=Optional[WorkflowWithoutID]
+    "/i/{image_name}/workflow", operation_id="get_image_workflow", response_model=WorkflowAndGraphResponse
 )
 async def get_image_workflow(
     image_name: str = Path(description="The name of image whose workflow to get"),
-) -> Optional[WorkflowWithoutID]:
+) -> WorkflowAndGraphResponse:
     try:
-        return ApiDependencies.invoker.services.images.get_workflow(image_name)
+        workflow = ApiDependencies.invoker.services.images.get_workflow(image_name)
+        graph = ApiDependencies.invoker.services.images.get_graph(image_name)
+        return WorkflowAndGraphResponse(workflow=workflow, graph=graph)
     except Exception:
         raise HTTPException(status_code=404)
 

invokeai/app/api/routers/model_manager.py

@@ -6,17 +6,18 @@ import pathlib
 import shutil
 import traceback
 from copy import deepcopy
-from typing import Any, Dict, List, Optional
+from typing import Any, Dict, List, Optional, Type
 
 from fastapi import Body, Path, Query, Response, UploadFile
-from fastapi.responses import FileResponse
+from fastapi.responses import FileResponse, HTMLResponse
 from fastapi.routing import APIRouter
 from PIL import Image
 from pydantic import AnyHttpUrl, BaseModel, ConfigDict, Field
 from starlette.exceptions import HTTPException
 from typing_extensions import Annotated
 
-from invokeai.app.services.model_install import ModelInstallJob
+from invokeai.app.services.model_images.model_images_common import ModelImageFileNotFoundException
+from invokeai.app.services.model_install.model_install_common import ModelInstallJob
 from invokeai.app.services.model_records import (
     DuplicateModelException,
     InvalidModelException,
@@ -52,6 +53,13 @@ class ModelsList(BaseModel):
     model_config = ConfigDict(use_enum_values=True)
 
 
+def add_cover_image_to_model_config(config: AnyModelConfig, dependencies: Type[ApiDependencies]) -> AnyModelConfig:
+    """Add a cover image URL to a model configuration."""
+    cover_image = dependencies.invoker.services.model_images.get_url(config.key)
+    config.cover_image = cover_image
+    return config
+
+
 ##############################################################################
 # These are example inputs and outputs that are used in places where Swagger
 # is unable to generate a correct example.
@@ -118,8 +126,7 @@ async def list_model_records(
             record_store.search_by_attr(model_type=model_type, model_name=model_name, model_format=model_format)
         )
     for model in found_models:
-        cover_image = ApiDependencies.invoker.services.model_images.get_url(model.key)
-        model.cover_image = cover_image
+        model = add_cover_image_to_model_config(model, ApiDependencies)
     return ModelsList(models=found_models)
 
 
@@ -160,12 +167,9 @@ async def get_model_record(
     key: str = Path(description="Key of the model record to fetch."),
 ) -> AnyModelConfig:
     """Get a model record"""
-    record_store = ApiDependencies.invoker.services.model_manager.store
     try:
-        config: AnyModelConfig = record_store.get_model(key)
-        cover_image = ApiDependencies.invoker.services.model_images.get_url(key)
-        config.cover_image = cover_image
-        return config
+        config = ApiDependencies.invoker.services.model_manager.store.get_model(key)
+        return add_cover_image_to_model_config(config, ApiDependencies)
     except UnknownModelException as e:
         raise HTTPException(status_code=404, detail=str(e))
 
@@ -294,14 +298,15 @@ async def update_model_record(
     installer = ApiDependencies.invoker.services.model_manager.install
     try:
         record_store.update_model(key, changes=changes)
-        model_response: AnyModelConfig = installer.sync_model_path(key)
+        config = installer.sync_model_path(key)
+        config = add_cover_image_to_model_config(config, ApiDependencies)
         logger.info(f"Updated model: {key}")
     except UnknownModelException as e:
         raise HTTPException(status_code=404, detail=str(e))
     except ValueError as e:
         logger.error(str(e))
         raise HTTPException(status_code=409, detail=str(e))
-    return model_response
+    return config
 
 
 @model_manager_router.get(
@@ -497,6 +502,133 @@ async def install_model(
     return result
 
 
+@model_manager_router.get(
+    "/install/huggingface",
+    operation_id="install_hugging_face_model",
+    responses={
+        201: {"description": "The model is being installed"},
+        400: {"description": "Bad request"},
+        409: {"description": "There is already a model corresponding to this path or repo_id"},
+    },
+    status_code=201,
+    response_class=HTMLResponse,
+)
+async def install_hugging_face_model(
+    source: str = Query(description="HuggingFace repo_id to install"),
+) -> HTMLResponse:
+    """Install a Hugging Face model using a string identifier."""
+
+    def generate_html(title: str, heading: str, repo_id: str, is_error: bool, message: str | None = "") -> str:
+        if message:
+            message = f"<p>{message}</p>"
+        title_class = "error" if is_error else "success"
+        return f"""
+            <html>
+
+            <head>
+                <title>{title}</title>
+                <style>
+                    body {{
+                        text-align: center;
+                        background-color: hsl(220 12% 10% / 1);
+                        font-family: Helvetica, sans-serif;
+                        color: hsl(220 12% 86% / 1);
+                    }}
+
+                    .repo-id {{
+                        color: hsl(220 12% 68% / 1);
+                    }}
+
+                    .error {{
+                        color: hsl(0 42% 68% / 1)
+                    }}
+
+                    .message-box {{
+                        display: inline-block;
+                        border-radius: 5px;
+                        background-color: hsl(220 12% 20% / 1);
+                        padding-inline-end: 30px;
+                        padding: 20px;
+                        padding-inline-start: 30px;
+                        padding-inline-end: 30px;
+                    }}
+
+                    .container {{
+                        display: flex;
+                        width: 100%;
+                        height: 100%;
+                        align-items: center;
+                        justify-content: center;
+                    }}
+
+                    a {{
+                        color: inherit
+                    }}
+
+                    a:visited {{
+                        color: inherit
+                    }}
+
+                    a:active {{
+                        color: inherit
+                    }}
+                </style>
+            </head>
+
+            <body style="background-color: hsl(220 12% 10% / 1);">
+                <div class="container">
+                    <div class="message-box">
+                        <h2 class="{title_class}">{heading}</h2>
+                        {message}
+                        <p class="repo-id">Repo ID: {repo_id}</p>
+                    </div>
+                </div>
+            </body>
+
+            </html>
+        """
+
+    try:
+        metadata = HuggingFaceMetadataFetch().from_id(source)
+        assert isinstance(metadata, ModelMetadataWithFiles)
+    except UnknownMetadataException:
+        title = "Unable to Install Model"
+        heading = "No HuggingFace repository found with that repo ID."
+        message = "Ensure the repo ID is correct and try again."
+        return HTMLResponse(content=generate_html(title, heading, source, True, message), status_code=400)
+
+    logger = ApiDependencies.invoker.services.logger
+
+    try:
+        installer = ApiDependencies.invoker.services.model_manager.install
+        if metadata.is_diffusers:
+            installer.heuristic_import(
+                source=source,
+                inplace=False,
+            )
+        elif metadata.ckpt_urls is not None and len(metadata.ckpt_urls) == 1:
+            installer.heuristic_import(
+                source=str(metadata.ckpt_urls[0]),
+                inplace=False,
+            )
+        else:
+            title = "Unable to Install Model"
+            heading = "This HuggingFace repo has multiple models."
+            message = "Please use the Model Manager to install this model."
+            return HTMLResponse(content=generate_html(title, heading, source, True, message), status_code=200)
+
+        title = "Model Install Started"
+        heading = "Your HuggingFace model is installing now."
+        message = "You can close this tab and check the Model Manager for installation progress."
+        return HTMLResponse(content=generate_html(title, heading, source, False, message), status_code=201)
+    except Exception as e:
+        logger.error(str(e))
+        title = "Unable to Install Model"
+        heading = "There was an problem installing this model."
+        message = 'Please use the Model Manager directly to install this model. If the issue persists, ask for help on <a href="https://discord.gg/ZmtBAhwWhy">discord</a>.'
+        return HTMLResponse(content=generate_html(title, heading, source, True, message), status_code=500)
+
+
 @model_manager_router.get(
     "/install",
     operation_id="list_model_installs",
@@ -648,6 +780,14 @@ async def convert_model(
         logger.error(str(e))
         raise HTTPException(status_code=409, detail=str(e))
 
+    # Update the model image if the model had one
+    try:
+        model_image = ApiDependencies.invoker.services.model_images.get(key)
+        ApiDependencies.invoker.services.model_images.save(model_image, new_key)
+        ApiDependencies.invoker.services.model_images.delete(key)
+    except ModelImageFileNotFoundException:
+        pass
+
     # delete the original safetensors file
     installer.delete(key)
 
@@ -655,7 +795,8 @@ async def convert_model(
     shutil.rmtree(cache_path)
 
     # return the config record for the new diffusers directory
-    new_config: AnyModelConfig = store.get_model(new_key)
+    new_config = store.get_model(new_key)
+    new_config = add_cover_image_to_model_config(new_config, ApiDependencies)
     return new_config
 
 

invokeai/app/api/routers/session_queue.py

@@ -203,6 +203,7 @@ async def get_batch_status(
     responses={
         200: {"model": SessionQueueItem},
     },
+    response_model_exclude_none=True,
 )
 async def get_queue_item(
     queue_id: str = Path(description="The queue id to perform this operation on"),

invokeai/app/api/sockets.py

@@ -1,66 +1,125 @@
 # Copyright (c) 2022 Kyle Schouviller (https://github.com/kyle0654)
 
+from typing import Any
+
 from fastapi import FastAPI
-from fastapi_events.handlers.local import local_handler
-from fastapi_events.typing import Event
+from pydantic import BaseModel
 from socketio import ASGIApp, AsyncServer
 
-from ..services.events.events_base import EventServiceBase
+from invokeai.app.services.events.events_common import (
+    BatchEnqueuedEvent,
+    BulkDownloadCompleteEvent,
+    BulkDownloadErrorEvent,
+    BulkDownloadEventBase,
+    BulkDownloadStartedEvent,
+    DownloadCancelledEvent,
+    DownloadCompleteEvent,
+    DownloadErrorEvent,
+    DownloadEventBase,
+    DownloadProgressEvent,
+    DownloadStartedEvent,
+    FastAPIEvent,
+    InvocationCompleteEvent,
+    InvocationDenoiseProgressEvent,
+    InvocationErrorEvent,
+    InvocationStartedEvent,
+    ModelEventBase,
+    ModelInstallCancelledEvent,
+    ModelInstallCompleteEvent,
+    ModelInstallDownloadProgressEvent,
+    ModelInstallDownloadsCompleteEvent,
+    ModelInstallErrorEvent,
+    ModelInstallStartedEvent,
+    ModelLoadCompleteEvent,
+    ModelLoadStartedEvent,
+    QueueClearedEvent,
+    QueueEventBase,
+    QueueItemStatusChangedEvent,
+    register_events,
+)
+
+
+class QueueSubscriptionEvent(BaseModel):
+    """Event data for subscribing to the socket.io queue room.
+    This is a pydantic model to ensure the data is in the correct format."""
+
+    queue_id: str
+
+
+class BulkDownloadSubscriptionEvent(BaseModel):
+    """Event data for subscribing to the socket.io bulk downloads room.
+    This is a pydantic model to ensure the data is in the correct format."""
+
+    bulk_download_id: str
+
+
+QUEUE_EVENTS = {
+    InvocationStartedEvent,
+    InvocationDenoiseProgressEvent,
+    InvocationCompleteEvent,
+    InvocationErrorEvent,
+    QueueItemStatusChangedEvent,
+    BatchEnqueuedEvent,
+    QueueClearedEvent,
+}
+
+MODEL_EVENTS = {
+    DownloadCancelledEvent,
+    DownloadCompleteEvent,
+    DownloadErrorEvent,
+    DownloadProgressEvent,
+    DownloadStartedEvent,
+    ModelLoadStartedEvent,
+    ModelLoadCompleteEvent,
+    ModelInstallDownloadProgressEvent,
+    ModelInstallDownloadsCompleteEvent,
+    ModelInstallStartedEvent,
+    ModelInstallCompleteEvent,
+    ModelInstallCancelledEvent,
+    ModelInstallErrorEvent,
+}
+
+BULK_DOWNLOAD_EVENTS = {BulkDownloadStartedEvent, BulkDownloadCompleteEvent, BulkDownloadErrorEvent}
 
 
 class SocketIO:
-    __sio: AsyncServer
-    __app: ASGIApp
+    _sub_queue = "subscribe_queue"
+    _unsub_queue = "unsubscribe_queue"
 
-    __sub_queue: str = "subscribe_queue"
-    __unsub_queue: str = "unsubscribe_queue"
-
-    __sub_bulk_download: str = "subscribe_bulk_download"
-    __unsub_bulk_download: str = "unsubscribe_bulk_download"
+    _sub_bulk_download = "subscribe_bulk_download"
+    _unsub_bulk_download = "unsubscribe_bulk_download"
 
     def __init__(self, app: FastAPI):
-        self.__sio = AsyncServer(async_mode="asgi", cors_allowed_origins="*")
-        self.__app = ASGIApp(socketio_server=self.__sio, socketio_path="/ws/socket.io")
-        app.mount("/ws", self.__app)
+        self._sio = AsyncServer(async_mode="asgi", cors_allowed_origins="*")
+        self._app = ASGIApp(socketio_server=self._sio, socketio_path="/ws/socket.io")
+        app.mount("/ws", self._app)
 
-        self.__sio.on(self.__sub_queue, handler=self._handle_sub_queue)
-        self.__sio.on(self.__unsub_queue, handler=self._handle_unsub_queue)
-        local_handler.register(event_name=EventServiceBase.queue_event, _func=self._handle_queue_event)
-        local_handler.register(event_name=EventServiceBase.model_event, _func=self._handle_model_event)
+        self._sio.on(self._sub_queue, handler=self._handle_sub_queue)
+        self._sio.on(self._unsub_queue, handler=self._handle_unsub_queue)
+        self._sio.on(self._sub_bulk_download, handler=self._handle_sub_bulk_download)
+        self._sio.on(self._unsub_bulk_download, handler=self._handle_unsub_bulk_download)
 
-        self.__sio.on(self.__sub_bulk_download, handler=self._handle_sub_bulk_download)
-        self.__sio.on(self.__unsub_bulk_download, handler=self._handle_unsub_bulk_download)
-        local_handler.register(event_name=EventServiceBase.bulk_download_event, _func=self._handle_bulk_download_event)
+        register_events(QUEUE_EVENTS, self._handle_queue_event)
+        register_events(MODEL_EVENTS, self._handle_model_event)
+        register_events(BULK_DOWNLOAD_EVENTS, self._handle_bulk_image_download_event)
 
-    async def _handle_queue_event(self, event: Event):
-        await self.__sio.emit(
-            event=event[1]["event"],
-            data=event[1]["data"],
-            room=event[1]["data"]["queue_id"],
-        )
+    async def _handle_sub_queue(self, sid: str, data: Any) -> None:
+        await self._sio.enter_room(sid, QueueSubscriptionEvent(**data).queue_id)
 
-    async def _handle_sub_queue(self, sid, data, *args, **kwargs) -> None:
-        if "queue_id" in data:
-            await self.__sio.enter_room(sid, data["queue_id"])
+    async def _handle_unsub_queue(self, sid: str, data: Any) -> None:
+        await self._sio.leave_room(sid, QueueSubscriptionEvent(**data).queue_id)
 
-    async def _handle_unsub_queue(self, sid, data, *args, **kwargs) -> None:
-        if "queue_id" in data:
-            await self.__sio.leave_room(sid, data["queue_id"])
+    async def _handle_sub_bulk_download(self, sid: str, data: Any) -> None:
+        await self._sio.enter_room(sid, BulkDownloadSubscriptionEvent(**data).bulk_download_id)
 
-    async def _handle_model_event(self, event: Event) -> None:
-        await self.__sio.emit(event=event[1]["event"], data=event[1]["data"])
+    async def _handle_unsub_bulk_download(self, sid: str, data: Any) -> None:
+        await self._sio.leave_room(sid, BulkDownloadSubscriptionEvent(**data).bulk_download_id)
 
-    async def _handle_bulk_download_event(self, event: Event):
-        await self.__sio.emit(
-            event=event[1]["event"],
-            data=event[1]["data"],
-            room=event[1]["data"]["bulk_download_id"],
-        )
+    async def _handle_queue_event(self, event: FastAPIEvent[QueueEventBase]):
+        await self._sio.emit(event=event[0], data=event[1].model_dump(mode="json"), room=event[1].queue_id)
 
-    async def _handle_sub_bulk_download(self, sid, data, *args, **kwargs):
-        if "bulk_download_id" in data:
-            await self.__sio.enter_room(sid, data["bulk_download_id"])
+    async def _handle_model_event(self, event: FastAPIEvent[ModelEventBase | DownloadEventBase]) -> None:
+        await self._sio.emit(event=event[0], data=event[1].model_dump(mode="json"))
 
-    async def _handle_unsub_bulk_download(self, sid, data, *args, **kwargs):
-        if "bulk_download_id" in data:
-            await self.__sio.leave_room(sid, data["bulk_download_id"])
+    async def _handle_bulk_image_download_event(self, event: FastAPIEvent[BulkDownloadEventBase]) -> None:
+        await self._sio.emit(event=event[0], data=event[1].model_dump(mode="json"), room=event[1].bulk_download_id)

invokeai/app/api_app.py

@@ -3,9 +3,7 @@ import logging
 import mimetypes
 import socket
 from contextlib import asynccontextmanager
-from inspect import signature
 from pathlib import Path
-from typing import Any
 
 import torch
 import uvicorn
@@ -13,11 +11,9 @@ from fastapi import FastAPI
 from fastapi.middleware.cors import CORSMiddleware
 from fastapi.middleware.gzip import GZipMiddleware
 from fastapi.openapi.docs import get_redoc_html, get_swagger_ui_html
-from fastapi.openapi.utils import get_openapi
 from fastapi.responses import HTMLResponse
 from fastapi_events.handlers.local import local_handler
 from fastapi_events.middleware import EventHandlerASGIMiddleware
-from pydantic.json_schema import models_json_schema
 from torch.backends.mps import is_available as is_mps_available
 
 # for PyCharm:
@@ -25,9 +21,8 @@ from torch.backends.mps import is_available as is_mps_available
 import invokeai.backend.util.hotfixes  # noqa: F401 (monkeypatching on import)
 import invokeai.frontend.web as web_dir
 from invokeai.app.api.no_cache_staticfiles import NoCacheStaticFiles
-from invokeai.app.invocations.model import ModelIdentifierField
 from invokeai.app.services.config.config_default import get_config
-from invokeai.app.services.session_processor.session_processor_common import ProgressImage
+from invokeai.app.util.custom_openapi import get_openapi_func
 from invokeai.backend.util.devices import TorchDevice
 
 from ..backend.util.logging import InvokeAILogger
@@ -44,11 +39,6 @@ from .api.routers import (
     workflows,
 )
 from .api.sockets import SocketIO
-from .invocations.baseinvocation import (
-    BaseInvocation,
-    UIConfigBase,
-)
-from .invocations.fields import InputFieldJSONSchemaExtra, OutputFieldJSONSchemaExtra
 
 app_config = get_config()
 
@@ -118,85 +108,7 @@ app.include_router(app_info.app_router, prefix="/api")
 app.include_router(session_queue.session_queue_router, prefix="/api")
 app.include_router(workflows.workflows_router, prefix="/api")
 
-
-# Build a custom OpenAPI to include all outputs
-# TODO: can outputs be included on metadata of invocation schemas somehow?
-def custom_openapi() -> dict[str, Any]:
-    if app.openapi_schema:
-        return app.openapi_schema
-    openapi_schema = get_openapi(
-        title=app.title,
-        description="An API for invoking AI image operations",
-        version="1.0.0",
-        routes=app.routes,
-        separate_input_output_schemas=False,  # https://fastapi.tiangolo.com/how-to/separate-openapi-schemas/
-    )
-
-    # Add all outputs
-    all_invocations = BaseInvocation.get_invocations()
-    output_types = set()
-    output_type_titles = {}
-    for invoker in all_invocations:
-        output_type = signature(invoker.invoke).return_annotation
-        output_types.add(output_type)
-
-    output_schemas = models_json_schema(
-        models=[(o, "serialization") for o in output_types], ref_template="#/components/schemas/{model}"
-    )
-    for schema_key, output_schema in output_schemas[1]["$defs"].items():
-        # TODO: note that we assume the schema_key here is the TYPE.__name__
-        # This could break in some cases, figure out a better way to do it
-        output_type_titles[schema_key] = output_schema["title"]
-        openapi_schema["components"]["schemas"][schema_key] = output_schema
-        openapi_schema["components"]["schemas"][schema_key]["class"] = "output"
-
-    # Some models don't end up in the schemas as standalone definitions
-    additional_schemas = models_json_schema(
-        [
-            (UIConfigBase, "serialization"),
-            (InputFieldJSONSchemaExtra, "serialization"),
-            (OutputFieldJSONSchemaExtra, "serialization"),
-            (ModelIdentifierField, "serialization"),
-            (ProgressImage, "serialization"),
-        ],
-        ref_template="#/components/schemas/{model}",
-    )
-    for schema_key, schema_json in additional_schemas[1]["$defs"].items():
-        openapi_schema["components"]["schemas"][schema_key] = schema_json
-
-    # Add a reference to the output type to additionalProperties of the invoker schema
-    for invoker in all_invocations:
-        invoker_name = invoker.__name__  # type: ignore [attr-defined] # this is a valid attribute
-        output_type = signature(obj=invoker.invoke).return_annotation
-        output_type_title = output_type_titles[output_type.__name__]
-        invoker_schema = openapi_schema["components"]["schemas"][f"{invoker_name}"]
-        outputs_ref = {"$ref": f"#/components/schemas/{output_type_title}"}
-        invoker_schema["output"] = outputs_ref
-        invoker_schema["class"] = "invocation"
-
-    # This code no longer seems to be necessary?
-    # Leave it here just in case
-    #
-    # from invokeai.backend.model_manager import get_model_config_formats
-    # formats = get_model_config_formats()
-    # for model_config_name, enum_set in formats.items():
-
-    #     if model_config_name in openapi_schema["components"]["schemas"]:
-    #         # print(f"Config with name {name} already defined")
-    #         continue
-
-    #     openapi_schema["components"]["schemas"][model_config_name] = {
-    #         "title": model_config_name,
-    #         "description": "An enumeration.",
-    #         "type": "string",
-    #         "enum": [v.value for v in enum_set],
-    #     }
-
-    app.openapi_schema = openapi_schema
-    return app.openapi_schema
-
-
-app.openapi = custom_openapi  # type: ignore [method-assign] # this is a valid assignment
+app.openapi = get_openapi_func(app)
 
 
 @app.get("/docs", include_in_schema=False)

invokeai/app/invocations/baseinvocation.py

@@ -98,11 +98,13 @@ class BaseInvocationOutput(BaseModel):
 
     _output_classes: ClassVar[set[BaseInvocationOutput]] = set()
     _typeadapter: ClassVar[Optional[TypeAdapter[Any]]] = None
+    _typeadapter_needs_update: ClassVar[bool] = False
 
     @classmethod
     def register_output(cls, output: BaseInvocationOutput) -> None:
         """Registers an invocation output."""
         cls._output_classes.add(output)
+        cls._typeadapter_needs_update = True
 
     @classmethod
     def get_outputs(cls) -> Iterable[BaseInvocationOutput]:
@@ -112,11 +114,12 @@ class BaseInvocationOutput(BaseModel):
     @classmethod
     def get_typeadapter(cls) -> TypeAdapter[Any]:
         """Gets a pydantc TypeAdapter for the union of all invocation output types."""
-        if not cls._typeadapter:
-            InvocationOutputsUnion = TypeAliasType(
-                "InvocationOutputsUnion", Annotated[Union[tuple(cls._output_classes)], Field(discriminator="type")]
+        if not cls._typeadapter or cls._typeadapter_needs_update:
+            AnyInvocationOutput = TypeAliasType(
+                "AnyInvocationOutput", Annotated[Union[tuple(cls._output_classes)], Field(discriminator="type")]
             )
-            cls._typeadapter = TypeAdapter(InvocationOutputsUnion)
+            cls._typeadapter = TypeAdapter(AnyInvocationOutput)
+            cls._typeadapter_needs_update = False
         return cls._typeadapter
 
     @classmethod
@@ -125,12 +128,13 @@ class BaseInvocationOutput(BaseModel):
         return (i.get_type() for i in BaseInvocationOutput.get_outputs())
 
     @staticmethod
-    def json_schema_extra(schema: dict[str, Any], model_class: Type[BaseModel]) -> None:
+    def json_schema_extra(schema: dict[str, Any], model_class: Type[BaseInvocationOutput]) -> None:
         """Adds various UI-facing attributes to the invocation output's OpenAPI schema."""
         # Because we use a pydantic Literal field with default value for the invocation type,
         # it will be typed as optional in the OpenAPI schema. Make it required manually.
         if "required" not in schema or not isinstance(schema["required"], list):
             schema["required"] = []
+        schema["class"] = "output"
         schema["required"].extend(["type"])
 
     @classmethod
@@ -167,6 +171,7 @@ class BaseInvocation(ABC, BaseModel):
 
     _invocation_classes: ClassVar[set[BaseInvocation]] = set()
     _typeadapter: ClassVar[Optional[TypeAdapter[Any]]] = None
+    _typeadapter_needs_update: ClassVar[bool] = False
 
     @classmethod
     def get_type(cls) -> str:
@@ -177,15 +182,17 @@ class BaseInvocation(ABC, BaseModel):
     def register_invocation(cls, invocation: BaseInvocation) -> None:
         """Registers an invocation."""
         cls._invocation_classes.add(invocation)
+        cls._typeadapter_needs_update = True
 
     @classmethod
     def get_typeadapter(cls) -> TypeAdapter[Any]:
         """Gets a pydantc TypeAdapter for the union of all invocation types."""
-        if not cls._typeadapter:
-            InvocationsUnion = TypeAliasType(
-                "InvocationsUnion", Annotated[Union[tuple(cls._invocation_classes)], Field(discriminator="type")]
+        if not cls._typeadapter or cls._typeadapter_needs_update:
+            AnyInvocation = TypeAliasType(
+                "AnyInvocation", Annotated[Union[tuple(cls._invocation_classes)], Field(discriminator="type")]
             )
-            cls._typeadapter = TypeAdapter(InvocationsUnion)
+            cls._typeadapter = TypeAdapter(AnyInvocation)
+            cls._typeadapter_needs_update = False
         return cls._typeadapter
 
     @classmethod
@@ -221,7 +228,7 @@ class BaseInvocation(ABC, BaseModel):
         return signature(cls.invoke).return_annotation
 
     @staticmethod
-    def json_schema_extra(schema: dict[str, Any], model_class: Type[BaseModel], *args, **kwargs) -> None:
+    def json_schema_extra(schema: dict[str, Any], model_class: Type[BaseInvocation]) -> None:
         """Adds various UI-facing attributes to the invocation's OpenAPI schema."""
         uiconfig = cast(UIConfigBase | None, getattr(model_class, "UIConfig", None))
         if uiconfig is not None:
@@ -237,6 +244,7 @@ class BaseInvocation(ABC, BaseModel):
             schema["version"] = uiconfig.version
         if "required" not in schema or not isinstance(schema["required"], list):
             schema["required"] = []
+        schema["class"] = "invocation"
         schema["required"].extend(["type", "id"])
 
     @abstractmethod
@@ -310,7 +318,7 @@ class BaseInvocation(ABC, BaseModel):
         protected_namespaces=(),
         validate_assignment=True,
         json_schema_extra=json_schema_extra,
-        json_schema_serialization_defaults_required=True,
+        json_schema_serialization_defaults_required=False,
         coerce_numbers_to_str=True,
     )
 

invokeai/app/invocations/blend_latents.py

@@ -0,0 +1,98 @@
+from typing import Any, Union
+
+import numpy as np
+import numpy.typing as npt
+import torch
+
+from invokeai.app.invocations.baseinvocation import BaseInvocation, invocation
+from invokeai.app.invocations.fields import FieldDescriptions, Input, InputField, LatentsField
+from invokeai.app.invocations.primitives import LatentsOutput
+from invokeai.app.services.shared.invocation_context import InvocationContext
+from invokeai.backend.util.devices import TorchDevice
+
+
+@invocation(
+    "lblend",
+    title="Blend Latents",
+    tags=["latents", "blend"],
+    category="latents",
+    version="1.0.3",
+)
+class BlendLatentsInvocation(BaseInvocation):
+    """Blend two latents using a given alpha. Latents must have same size."""
+
+    latents_a: LatentsField = InputField(
+        description=FieldDescriptions.latents,
+        input=Input.Connection,
+    )
+    latents_b: LatentsField = InputField(
+        description=FieldDescriptions.latents,
+        input=Input.Connection,
+    )
+    alpha: float = InputField(default=0.5, description=FieldDescriptions.blend_alpha)
+
+    def invoke(self, context: InvocationContext) -> LatentsOutput:
+        latents_a = context.tensors.load(self.latents_a.latents_name)
+        latents_b = context.tensors.load(self.latents_b.latents_name)
+
+        if latents_a.shape != latents_b.shape:
+            raise Exception("Latents to blend must be the same size.")
+
+        device = TorchDevice.choose_torch_device()
+
+        def slerp(
+            t: Union[float, npt.NDArray[Any]],  # FIXME: maybe use np.float32 here?
+            v0: Union[torch.Tensor, npt.NDArray[Any]],
+            v1: Union[torch.Tensor, npt.NDArray[Any]],
+            DOT_THRESHOLD: float = 0.9995,
+        ) -> Union[torch.Tensor, npt.NDArray[Any]]:
+            """
+            Spherical linear interpolation
+            Args:
+                t (float/np.ndarray): Float value between 0.0 and 1.0
+                v0 (np.ndarray): Starting vector
+                v1 (np.ndarray): Final vector
+                DOT_THRESHOLD (float): Threshold for considering the two vectors as
+                                    colineal. Not recommended to alter this.
+            Returns:
+                v2 (np.ndarray): Interpolation vector between v0 and v1
+            """
+            inputs_are_torch = False
+            if not isinstance(v0, np.ndarray):
+                inputs_are_torch = True
+                v0 = v0.detach().cpu().numpy()
+            if not isinstance(v1, np.ndarray):
+                inputs_are_torch = True
+                v1 = v1.detach().cpu().numpy()
+
+            dot = np.sum(v0 * v1 / (np.linalg.norm(v0) * np.linalg.norm(v1)))
+            if np.abs(dot) > DOT_THRESHOLD:
+                v2 = (1 - t) * v0 + t * v1
+            else:
+                theta_0 = np.arccos(dot)
+                sin_theta_0 = np.sin(theta_0)
+                theta_t = theta_0 * t
+                sin_theta_t = np.sin(theta_t)
+                s0 = np.sin(theta_0 - theta_t) / sin_theta_0
+                s1 = sin_theta_t / sin_theta_0
+                v2 = s0 * v0 + s1 * v1
+
+            if inputs_are_torch:
+                v2_torch: torch.Tensor = torch.from_numpy(v2).to(device)
+                return v2_torch
+            else:
+                assert isinstance(v2, np.ndarray)
+                return v2
+
+        # blend
+        bl = slerp(self.alpha, latents_a, latents_b)
+        assert isinstance(bl, torch.Tensor)
+        blended_latents: torch.Tensor = bl  # for type checking convenience
+
+        # https://discuss.huggingface.co/t/memory-usage-by-later-pipeline-stages/23699
+        blended_latents = blended_latents.to("cpu")
+
+        TorchDevice.empty_cache()
+
+        name = context.tensors.save(tensor=blended_latents)
+        return LatentsOutput.build(latents_name=name, latents=blended_latents, seed=self.latents_a.seed)

invokeai/app/invocations/compel.py

@@ -65,11 +65,7 @@ class CompelInvocation(BaseInvocation):
     @torch.no_grad()
     def invoke(self, context: InvocationContext) -> ConditioningOutput:
         tokenizer_info = context.models.load(self.clip.tokenizer)
-        tokenizer_model = tokenizer_info.model
-        assert isinstance(tokenizer_model, CLIPTokenizer)
         text_encoder_info = context.models.load(self.clip.text_encoder)
-        text_encoder_model = text_encoder_info.model
-        assert isinstance(text_encoder_model, CLIPTextModel)
 
         def _lora_loader() -> Iterator[Tuple[LoRAModelRaw, float]]:
             for lora in self.clip.loras:
@@ -84,29 +80,36 @@ class CompelInvocation(BaseInvocation):
         ti_list = generate_ti_list(self.prompt, text_encoder_info.config.base, context)
 
         with (
-            ModelPatcher.apply_ti(tokenizer_model, text_encoder_model, ti_list) as (
-                tokenizer,
+            # apply all patches while the model is on the target device
+            text_encoder_info.model_on_device() as (model_state_dict, text_encoder),
+            tokenizer_info as tokenizer,
+            ModelPatcher.apply_lora_text_encoder(
+                text_encoder,
+                loras=_lora_loader(),
+                model_state_dict=model_state_dict,
+            ),
+            # Apply CLIP Skip after LoRA to prevent LoRA application from failing on skipped layers.
+            ModelPatcher.apply_clip_skip(text_encoder, self.clip.skipped_layers),
+            ModelPatcher.apply_ti(tokenizer, text_encoder, ti_list) as (
+                patched_tokenizer,
                 ti_manager,
             ),
-            text_encoder_info as text_encoder,
-            # Apply the LoRA after text_encoder has been moved to its target device for faster patching.
-            ModelPatcher.apply_lora_text_encoder(text_encoder, _lora_loader()),
-            # Apply CLIP Skip after LoRA to prevent LoRA application from failing on skipped layers.
-            ModelPatcher.apply_clip_skip(text_encoder_model, self.clip.skipped_layers),
         ):
             assert isinstance(text_encoder, CLIPTextModel)
+            assert isinstance(tokenizer, CLIPTokenizer)
             compel = Compel(
-                tokenizer=tokenizer,
+                tokenizer=patched_tokenizer,
                 text_encoder=text_encoder,
                 textual_inversion_manager=ti_manager,
                 dtype_for_device_getter=TorchDevice.choose_torch_dtype,
                 truncate_long_prompts=False,
+                device=TorchDevice.choose_torch_device(),
             )
 
             conjunction = Compel.parse_prompt_string(self.prompt)
 
             if context.config.get().log_tokenization:
-                log_tokenization_for_conjunction(conjunction, tokenizer)
+                log_tokenization_for_conjunction(conjunction, patched_tokenizer)
 
             c, _options = compel.build_conditioning_tensor_for_conjunction(conjunction)
 
@@ -115,6 +118,7 @@ class CompelInvocation(BaseInvocation):
         conditioning_data = ConditioningFieldData(conditionings=[BasicConditioningInfo(embeds=c)])
 
         conditioning_name = context.conditioning.save(conditioning_data)
+
         return ConditioningOutput(
             conditioning=ConditioningField(
                 conditioning_name=conditioning_name,
@@ -136,11 +140,7 @@ class SDXLPromptInvocationBase:
         zero_on_empty: bool,
     ) -> Tuple[torch.Tensor, Optional[torch.Tensor]]:
         tokenizer_info = context.models.load(clip_field.tokenizer)
-        tokenizer_model = tokenizer_info.model
-        assert isinstance(tokenizer_model, CLIPTokenizer)
         text_encoder_info = context.models.load(clip_field.text_encoder)
-        text_encoder_model = text_encoder_info.model
-        assert isinstance(text_encoder_model, (CLIPTextModel, CLIPTextModelWithProjection))
 
         # return zero on empty
         if prompt == "" and zero_on_empty:
@@ -177,33 +177,42 @@ class SDXLPromptInvocationBase:
         ti_list = generate_ti_list(prompt, text_encoder_info.config.base, context)
 
         with (
-            ModelPatcher.apply_ti(tokenizer_model, text_encoder_model, ti_list) as (
-                tokenizer,
+            # apply all patches while the model is on the target device
+            text_encoder_info.model_on_device() as (state_dict, text_encoder),
+            tokenizer_info as tokenizer,
+            ModelPatcher.apply_lora(
+                text_encoder,
+                loras=_lora_loader(),
+                prefix=lora_prefix,
+                model_state_dict=state_dict,
+            ),
+            # Apply CLIP Skip after LoRA to prevent LoRA application from failing on skipped layers.
+            ModelPatcher.apply_clip_skip(text_encoder, clip_field.skipped_layers),
+            ModelPatcher.apply_ti(tokenizer, text_encoder, ti_list) as (
+                patched_tokenizer,
                 ti_manager,
             ),
-            text_encoder_info as text_encoder,
-            # Apply the LoRA after text_encoder has been moved to its target device for faster patching.
-            ModelPatcher.apply_lora(text_encoder, _lora_loader(), lora_prefix),
-            # Apply CLIP Skip after LoRA to prevent LoRA application from failing on skipped layers.
-            ModelPatcher.apply_clip_skip(text_encoder_model, clip_field.skipped_layers),
         ):
             assert isinstance(text_encoder, (CLIPTextModel, CLIPTextModelWithProjection))
+            assert isinstance(tokenizer, CLIPTokenizer)
+
             text_encoder = cast(CLIPTextModel, text_encoder)
             compel = Compel(
-                tokenizer=tokenizer,
+                tokenizer=patched_tokenizer,
                 text_encoder=text_encoder,
                 textual_inversion_manager=ti_manager,
                 dtype_for_device_getter=TorchDevice.choose_torch_dtype,
                 truncate_long_prompts=False,  # TODO:
                 returned_embeddings_type=ReturnedEmbeddingsType.PENULTIMATE_HIDDEN_STATES_NON_NORMALIZED,  # TODO: clip skip
                 requires_pooled=get_pooled,
+                device=TorchDevice.choose_torch_device(),
             )
 
             conjunction = Compel.parse_prompt_string(prompt)
 
             if context.config.get().log_tokenization:
                 # TODO: better logging for and syntax
-                log_tokenization_for_conjunction(conjunction, tokenizer)
+                log_tokenization_for_conjunction(conjunction, patched_tokenizer)
 
             # TODO: ask for optimizations? to not run text_encoder twice
             c, _options = compel.build_conditioning_tensor_for_conjunction(conjunction)
@@ -307,7 +316,6 @@ class SDXLCompelPromptInvocation(BaseInvocation, SDXLPromptInvocationBase):
                 )
             ]
         )
-
         conditioning_name = context.conditioning.save(conditioning_data)
 
         return ConditioningOutput(

invokeai/app/invocations/constants.py

@@ -1,6 +1,7 @@
 from typing import Literal
 
 from invokeai.backend.stable_diffusion.schedulers import SCHEDULER_MAP
+from invokeai.backend.util.devices import TorchDevice
 
 LATENT_SCALE_FACTOR = 8
 """
@@ -15,3 +16,5 @@ SCHEDULER_NAME_VALUES = Literal[tuple(SCHEDULER_MAP.keys())]
 
 IMAGE_MODES = Literal["L", "RGB", "RGBA", "CMYK", "YCbCr", "LAB", "HSV", "I", "F"]
 """A literal type for PIL image modes supported by Invoke"""
+
+DEFAULT_PRECISION = TorchDevice.choose_torch_dtype()

invokeai/app/invocations/controlnet_image_processors.py

@@ -2,6 +2,7 @@
 # initial implementation by Gregg Helt, 2023
 # heavily leverages controlnet_aux package: https://github.com/patrickvonplaten/controlnet_aux
 from builtins import bool, float
+from pathlib import Path
 from typing import Dict, List, Literal, Union
 
 import cv2
@@ -24,7 +25,6 @@ from pydantic import BaseModel, Field, field_validator, model_validator
 from invokeai.app.invocations.fields import (
     FieldDescriptions,
     ImageField,
-    Input,
     InputField,
     OutputField,
     UIType,
@@ -37,12 +37,13 @@ from invokeai.app.invocations.util import validate_begin_end_step, validate_weig
 from invokeai.app.services.shared.invocation_context import InvocationContext
 from invokeai.app.util.controlnet_utils import CONTROLNET_MODE_VALUES, CONTROLNET_RESIZE_VALUES, heuristic_resize
 from invokeai.backend.image_util.canny import get_canny_edges
-from invokeai.backend.image_util.depth_anything import DepthAnythingDetector
-from invokeai.backend.image_util.dw_openpose import DWOpenposeDetector
+from invokeai.backend.image_util.depth_anything import DEPTH_ANYTHING_MODELS, DepthAnythingDetector
+from invokeai.backend.image_util.dw_openpose import DWPOSE_MODELS, DWOpenposeDetector
 from invokeai.backend.image_util.hed import HEDProcessor
 from invokeai.backend.image_util.lineart import LineartProcessor
 from invokeai.backend.image_util.lineart_anime import LineartAnimeProcessor
 from invokeai.backend.image_util.util import np_to_pil, pil_to_np
+from invokeai.backend.util.devices import TorchDevice
 
 from .baseinvocation import BaseInvocation, BaseInvocationOutput, Classification, invocation, invocation_output
 
@@ -80,13 +81,13 @@ class ControlOutput(BaseInvocationOutput):
     control: ControlField = OutputField(description=FieldDescriptions.control)
 
 
-@invocation("controlnet", title="ControlNet", tags=["controlnet"], category="controlnet", version="1.1.1")
+@invocation("controlnet", title="ControlNet", tags=["controlnet"], category="controlnet", version="1.1.2")
 class ControlNetInvocation(BaseInvocation):
     """Collects ControlNet info to pass to other nodes"""
 
     image: ImageField = InputField(description="The control image")
     control_model: ModelIdentifierField = InputField(
-        description=FieldDescriptions.controlnet_model, input=Input.Direct, ui_type=UIType.ControlNetModel
+        description=FieldDescriptions.controlnet_model, ui_type=UIType.ControlNetModel
     )
     control_weight: Union[float, List[float]] = InputField(
         default=1.0, ge=-1, le=2, description="The weight given to the ControlNet"
@@ -140,6 +141,7 @@ class ImageProcessorInvocation(BaseInvocation, WithMetadata, WithBoard):
         return context.images.get_pil(self.image.image_name, "RGB")
 
     def invoke(self, context: InvocationContext) -> ImageOutput:
+        self._context = context
         raw_image = self.load_image(context)
         # image type should be PIL.PngImagePlugin.PngImageFile ?
         processed_image = self.run_processor(raw_image)
@@ -285,7 +287,8 @@ class MidasDepthImageProcessorInvocation(ImageProcessorInvocation):
     # depth_and_normal not supported in controlnet_aux v0.0.3
     # depth_and_normal: bool = InputField(default=False, description="whether to use depth and normal mode")
 
-    def run_processor(self, image):
+    def run_processor(self, image: Image.Image) -> Image.Image:
+        # TODO: replace from_pretrained() calls with context.models.download_and_cache() (or similar)
         midas_processor = MidasDetector.from_pretrained("lllyasviel/Annotators")
         processed_image = midas_processor(
             image,
@@ -312,7 +315,7 @@ class NormalbaeImageProcessorInvocation(ImageProcessorInvocation):
     detect_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.detect_res)
     image_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.image_res)
 
-    def run_processor(self, image):
+    def run_processor(self, image: Image.Image) -> Image.Image:
         normalbae_processor = NormalBaeDetector.from_pretrained("lllyasviel/Annotators")
         processed_image = normalbae_processor(
             image, detect_resolution=self.detect_resolution, image_resolution=self.image_resolution
@@ -331,7 +334,7 @@ class MlsdImageProcessorInvocation(ImageProcessorInvocation):
     thr_v: float = InputField(default=0.1, ge=0, description="MLSD parameter `thr_v`")
     thr_d: float = InputField(default=0.1, ge=0, description="MLSD parameter `thr_d`")
 
-    def run_processor(self, image):
+    def run_processor(self, image: Image.Image) -> Image.Image:
         mlsd_processor = MLSDdetector.from_pretrained("lllyasviel/Annotators")
         processed_image = mlsd_processor(
             image,
@@ -354,7 +357,7 @@ class PidiImageProcessorInvocation(ImageProcessorInvocation):
     safe: bool = InputField(default=False, description=FieldDescriptions.safe_mode)
     scribble: bool = InputField(default=False, description=FieldDescriptions.scribble_mode)
 
-    def run_processor(self, image):
+    def run_processor(self, image: Image.Image) -> Image.Image:
         pidi_processor = PidiNetDetector.from_pretrained("lllyasviel/Annotators")
         processed_image = pidi_processor(
             image,
@@ -382,7 +385,7 @@ class ContentShuffleImageProcessorInvocation(ImageProcessorInvocation):
     w: int = InputField(default=512, ge=0, description="Content shuffle `w` parameter")
     f: int = InputField(default=256, ge=0, description="Content shuffle `f` parameter")
 
-    def run_processor(self, image):
+    def run_processor(self, image: Image.Image) -> Image.Image:
         content_shuffle_processor = ContentShuffleDetector()
         processed_image = content_shuffle_processor(
             image,
@@ -406,7 +409,7 @@ class ContentShuffleImageProcessorInvocation(ImageProcessorInvocation):
 class ZoeDepthImageProcessorInvocation(ImageProcessorInvocation):
     """Applies Zoe depth processing to image"""
 
-    def run_processor(self, image):
+    def run_processor(self, image: Image.Image) -> Image.Image:
         zoe_depth_processor = ZoeDetector.from_pretrained("lllyasviel/Annotators")
         processed_image = zoe_depth_processor(image)
         return processed_image
@@ -427,7 +430,7 @@ class MediapipeFaceProcessorInvocation(ImageProcessorInvocation):
     detect_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.detect_res)
     image_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.image_res)
 
-    def run_processor(self, image):
+    def run_processor(self, image: Image.Image) -> Image.Image:
         mediapipe_face_processor = MediapipeFaceDetector()
         processed_image = mediapipe_face_processor(
             image,
@@ -455,7 +458,7 @@ class LeresImageProcessorInvocation(ImageProcessorInvocation):
     detect_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.detect_res)
     image_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.image_res)
 
-    def run_processor(self, image):
+    def run_processor(self, image: Image.Image) -> Image.Image:
         leres_processor = LeresDetector.from_pretrained("lllyasviel/Annotators")
         processed_image = leres_processor(
             image,
@@ -497,8 +500,8 @@ class TileResamplerProcessorInvocation(ImageProcessorInvocation):
         np_img = cv2.resize(np_img, (W, H), interpolation=cv2.INTER_AREA)
         return np_img
 
-    def run_processor(self, img):
-        np_img = np.array(img, dtype=np.uint8)
+    def run_processor(self, image: Image.Image) -> Image.Image:
+        np_img = np.array(image, dtype=np.uint8)
         processed_np_image = self.tile_resample(
             np_img,
             # res=self.tile_size,
@@ -521,7 +524,7 @@ class SegmentAnythingProcessorInvocation(ImageProcessorInvocation):
     detect_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.detect_res)
     image_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.image_res)
 
-    def run_processor(self, image):
+    def run_processor(self, image: Image.Image) -> Image.Image:
         # segment_anything_processor = SamDetector.from_pretrained("ybelkada/segment-anything", subfolder="checkpoints")
         segment_anything_processor = SamDetectorReproducibleColors.from_pretrained(
             "ybelkada/segment-anything", subfolder="checkpoints"
@@ -567,7 +570,7 @@ class ColorMapImageProcessorInvocation(ImageProcessorInvocation):
 
     color_map_tile_size: int = InputField(default=64, ge=1, description=FieldDescriptions.tile_size)
 
-    def run_processor(self, image: Image.Image):
+    def run_processor(self, image: Image.Image) -> Image.Image:
         np_image = np.array(image, dtype=np.uint8)
         height, width = np_image.shape[:2]
 
@@ -602,12 +605,18 @@ class DepthAnythingImageProcessorInvocation(ImageProcessorInvocation):
     )
     resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.image_res)
 
-    def run_processor(self, image: Image.Image):
-        depth_anything_detector = DepthAnythingDetector()
-        depth_anything_detector.load_model(model_size=self.model_size)
+    def run_processor(self, image: Image.Image) -> Image.Image:
+        def loader(model_path: Path):
+            return DepthAnythingDetector.load_model(
+                model_path, model_size=self.model_size, device=TorchDevice.choose_torch_device()
+            )
 
-        processed_image = depth_anything_detector(image=image, resolution=self.resolution)
-        return processed_image
+        with self._context.models.load_remote_model(
+            source=DEPTH_ANYTHING_MODELS[self.model_size], loader=loader
+        ) as model:
+            depth_anything_detector = DepthAnythingDetector(model, TorchDevice.choose_torch_device())
+            processed_image = depth_anything_detector(image=image, resolution=self.resolution)
+            return processed_image
 
 
 @invocation(
@@ -625,8 +634,11 @@ class DWOpenposeImageProcessorInvocation(ImageProcessorInvocation):
     draw_hands: bool = InputField(default=False)
     image_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.image_res)
 
-    def run_processor(self, image: Image.Image):
-        dw_openpose = DWOpenposeDetector()
+    def run_processor(self, image: Image.Image) -> Image.Image:
+        onnx_det = self._context.models.download_and_cache_model(DWPOSE_MODELS["yolox_l.onnx"])
+        onnx_pose = self._context.models.download_and_cache_model(DWPOSE_MODELS["dw-ll_ucoco_384.onnx"])
+
+        dw_openpose = DWOpenposeDetector(onnx_det=onnx_det, onnx_pose=onnx_pose)
         processed_image = dw_openpose(
             image,
             draw_face=self.draw_face,

invokeai/app/invocations/create_denoise_mask.py

@@ -0,0 +1,80 @@
+from typing import Optional
+
+import torch
+import torchvision.transforms as T
+from PIL import Image
+from torchvision.transforms.functional import resize as tv_resize
+
+from invokeai.app.invocations.baseinvocation import BaseInvocation, invocation
+from invokeai.app.invocations.constants import DEFAULT_PRECISION
+from invokeai.app.invocations.fields import FieldDescriptions, ImageField, Input, InputField
+from invokeai.app.invocations.image_to_latents import ImageToLatentsInvocation
+from invokeai.app.invocations.model import VAEField
+from invokeai.app.invocations.primitives import DenoiseMaskOutput
+from invokeai.app.services.shared.invocation_context import InvocationContext
+from invokeai.backend.stable_diffusion.diffusers_pipeline import image_resized_to_grid_as_tensor
+
+
+@invocation(
+    "create_denoise_mask",
+    title="Create Denoise Mask",
+    tags=["mask", "denoise"],
+    category="latents",
+    version="1.0.2",
+)
+class CreateDenoiseMaskInvocation(BaseInvocation):
+    """Creates mask for denoising model run."""
+
+    vae: VAEField = InputField(description=FieldDescriptions.vae, input=Input.Connection, ui_order=0)
+    image: Optional[ImageField] = InputField(default=None, description="Image which will be masked", ui_order=1)
+    mask: ImageField = InputField(description="The mask to use when pasting", ui_order=2)
+    tiled: bool = InputField(default=False, description=FieldDescriptions.tiled, ui_order=3)
+    fp32: bool = InputField(
+        default=DEFAULT_PRECISION == torch.float32,
+        description=FieldDescriptions.fp32,
+        ui_order=4,
+    )
+
+    def prep_mask_tensor(self, mask_image: Image.Image) -> torch.Tensor:
+        if mask_image.mode != "L":
+            mask_image = mask_image.convert("L")
+        mask_tensor: torch.Tensor = image_resized_to_grid_as_tensor(mask_image, normalize=False)
+        if mask_tensor.dim() == 3:
+            mask_tensor = mask_tensor.unsqueeze(0)
+        # if shape is not None:
+        #    mask_tensor = tv_resize(mask_tensor, shape, T.InterpolationMode.BILINEAR)
+        return mask_tensor
+
+    @torch.no_grad()
+    def invoke(self, context: InvocationContext) -> DenoiseMaskOutput:
+        if self.image is not None:
+            image = context.images.get_pil(self.image.image_name)
+            image_tensor = image_resized_to_grid_as_tensor(image.convert("RGB"))
+            if image_tensor.dim() == 3:
+                image_tensor = image_tensor.unsqueeze(0)
+        else:
+            image_tensor = None
+
+        mask = self.prep_mask_tensor(
+            context.images.get_pil(self.mask.image_name),
+        )
+
+        if image_tensor is not None:
+            vae_info = context.models.load(self.vae.vae)
+
+            img_mask = tv_resize(mask, image_tensor.shape[-2:], T.InterpolationMode.BILINEAR, antialias=False)
+            masked_image = image_tensor * torch.where(img_mask < 0.5, 0.0, 1.0)
+            # TODO:
+            masked_latents = ImageToLatentsInvocation.vae_encode(vae_info, self.fp32, self.tiled, masked_image.clone())
+
+            masked_latents_name = context.tensors.save(tensor=masked_latents)
+        else:
+            masked_latents_name = None
+
+        mask_name = context.tensors.save(tensor=mask)
+
+        return DenoiseMaskOutput.build(
+            mask_name=mask_name,
+            masked_latents_name=masked_latents_name,
+            gradient=False,
+        )

invokeai/app/invocations/create_gradient_mask.py

@@ -0,0 +1,138 @@
+from typing import Literal, Optional
+
+import numpy as np
+import torch
+import torchvision.transforms as T
+from PIL import Image, ImageFilter
+from torchvision.transforms.functional import resize as tv_resize
+
+from invokeai.app.invocations.baseinvocation import BaseInvocation, BaseInvocationOutput, invocation, invocation_output
+from invokeai.app.invocations.constants import DEFAULT_PRECISION
+from invokeai.app.invocations.fields import (
+    DenoiseMaskField,
+    FieldDescriptions,
+    ImageField,
+    Input,
+    InputField,
+    OutputField,
+)
+from invokeai.app.invocations.image_to_latents import ImageToLatentsInvocation
+from invokeai.app.invocations.model import UNetField, VAEField
+from invokeai.app.services.shared.invocation_context import InvocationContext
+from invokeai.backend.model_manager import LoadedModel
+from invokeai.backend.model_manager.config import MainConfigBase, ModelVariantType
+from invokeai.backend.stable_diffusion.diffusers_pipeline import image_resized_to_grid_as_tensor
+
+
+@invocation_output("gradient_mask_output")
+class GradientMaskOutput(BaseInvocationOutput):
+    """Outputs a denoise mask and an image representing the total gradient of the mask."""
+
+    denoise_mask: DenoiseMaskField = OutputField(description="Mask for denoise model run")
+    expanded_mask_area: ImageField = OutputField(
+        description="Image representing the total gradient area of the mask. For paste-back purposes."
+    )
+
+
+@invocation(
+    "create_gradient_mask",
+    title="Create Gradient Mask",
+    tags=["mask", "denoise"],
+    category="latents",
+    version="1.1.0",
+)
+class CreateGradientMaskInvocation(BaseInvocation):
+    """Creates mask for denoising model run."""
+
+    mask: ImageField = InputField(default=None, description="Image which will be masked", ui_order=1)
+    edge_radius: int = InputField(
+        default=16, ge=0, description="How far to blur/expand the edges of the mask", ui_order=2
+    )
+    coherence_mode: Literal["Gaussian Blur", "Box Blur", "Staged"] = InputField(default="Gaussian Blur", ui_order=3)
+    minimum_denoise: float = InputField(
+        default=0.0, ge=0, le=1, description="Minimum denoise level for the coherence region", ui_order=4
+    )
+    image: Optional[ImageField] = InputField(
+        default=None,
+        description="OPTIONAL: Only connect for specialized Inpainting models, masked_latents will be generated from the image with the VAE",
+        title="[OPTIONAL] Image",
+        ui_order=6,
+    )
+    unet: Optional[UNetField] = InputField(
+        description="OPTIONAL: If the Unet is a specialized Inpainting model, masked_latents will be generated from the image with the VAE",
+        default=None,
+        input=Input.Connection,
+        title="[OPTIONAL] UNet",
+        ui_order=5,
+    )
+    vae: Optional[VAEField] = InputField(
+        default=None,
+        description="OPTIONAL: Only connect for specialized Inpainting models, masked_latents will be generated from the image with the VAE",
+        title="[OPTIONAL] VAE",
+        input=Input.Connection,
+        ui_order=7,
+    )
+    tiled: bool = InputField(default=False, description=FieldDescriptions.tiled, ui_order=8)
+    fp32: bool = InputField(
+        default=DEFAULT_PRECISION == torch.float32,
+        description=FieldDescriptions.fp32,
+        ui_order=9,
+    )
+
+    @torch.no_grad()
+    def invoke(self, context: InvocationContext) -> GradientMaskOutput:
+        mask_image = context.images.get_pil(self.mask.image_name, mode="L")
+        if self.edge_radius > 0:
+            if self.coherence_mode == "Box Blur":
+                blur_mask = mask_image.filter(ImageFilter.BoxBlur(self.edge_radius))
+            else:  # Gaussian Blur OR Staged
+                # Gaussian Blur uses standard deviation. 1/2 radius is a good approximation
+                blur_mask = mask_image.filter(ImageFilter.GaussianBlur(self.edge_radius / 2))
+
+            blur_tensor: torch.Tensor = image_resized_to_grid_as_tensor(blur_mask, normalize=False)
+
+            # redistribute blur so that the original edges are 0 and blur outwards to 1
+            blur_tensor = (blur_tensor - 0.5) * 2
+
+            threshold = 1 - self.minimum_denoise
+
+            if self.coherence_mode == "Staged":
+                # wherever the blur_tensor is less than fully masked, convert it to threshold
+                blur_tensor = torch.where((blur_tensor < 1) & (blur_tensor > 0), threshold, blur_tensor)
+            else:
+                # wherever the blur_tensor is above threshold but less than 1, drop it to threshold
+                blur_tensor = torch.where((blur_tensor > threshold) & (blur_tensor < 1), threshold, blur_tensor)
+
+        else:
+            blur_tensor: torch.Tensor = image_resized_to_grid_as_tensor(mask_image, normalize=False)
+
+        mask_name = context.tensors.save(tensor=blur_tensor.unsqueeze(1))
+
+        # compute a [0, 1] mask from the blur_tensor
+        expanded_mask = torch.where((blur_tensor < 1), 0, 1)
+        expanded_mask_image = Image.fromarray((expanded_mask.squeeze(0).numpy() * 255).astype(np.uint8), mode="L")
+        expanded_image_dto = context.images.save(expanded_mask_image)
+
+        masked_latents_name = None
+        if self.unet is not None and self.vae is not None and self.image is not None:
+            # all three fields must be present at the same time
+            main_model_config = context.models.get_config(self.unet.unet.key)
+            assert isinstance(main_model_config, MainConfigBase)
+            if main_model_config.variant is ModelVariantType.Inpaint:
+                mask = blur_tensor
+                vae_info: LoadedModel = context.models.load(self.vae.vae)
+                image = context.images.get_pil(self.image.image_name)
+                image_tensor = image_resized_to_grid_as_tensor(image.convert("RGB"))
+                if image_tensor.dim() == 3:
+                    image_tensor = image_tensor.unsqueeze(0)
+                img_mask = tv_resize(mask, image_tensor.shape[-2:], T.InterpolationMode.BILINEAR, antialias=False)
+                masked_image = image_tensor * torch.where(img_mask < 0.5, 0.0, 1.0)
+                masked_latents = ImageToLatentsInvocation.vae_encode(
+                    vae_info, self.fp32, self.tiled, masked_image.clone()
+                )
+                masked_latents_name = context.tensors.save(tensor=masked_latents)
+
+        return GradientMaskOutput(
+            denoise_mask=DenoiseMaskField(mask_name=mask_name, masked_latents_name=masked_latents_name, gradient=True),
+            expanded_mask_area=ImageField(image_name=expanded_image_dto.image_name),
+        )

invokeai/app/invocations/crop_latents.py

@@ -0,0 +1,61 @@
+from invokeai.app.invocations.baseinvocation import BaseInvocation, invocation
+from invokeai.app.invocations.constants import LATENT_SCALE_FACTOR
+from invokeai.app.invocations.fields import FieldDescriptions, Input, InputField, LatentsField
+from invokeai.app.invocations.primitives import LatentsOutput
+from invokeai.app.services.shared.invocation_context import InvocationContext
+
+
+# The Crop Latents node was copied from @skunkworxdark's implementation here:
+# https://github.com/skunkworxdark/XYGrid_nodes/blob/74647fa9c1fa57d317a94bd43ca689af7f0aae5e/images_to_grids.py#L1117C1-L1167C80
+@invocation(
+    "crop_latents",
+    title="Crop Latents",
+    tags=["latents", "crop"],
+    category="latents",
+    version="1.0.2",
+)
+# TODO(ryand): Named `CropLatentsCoreInvocation` to prevent a conflict with custom node `CropLatentsInvocation`.
+# Currently, if the class names conflict then 'GET /openapi.json' fails.
+class CropLatentsCoreInvocation(BaseInvocation):
+    """Crops a latent-space tensor to a box specified in image-space. The box dimensions and coordinates must be
+    divisible by the latent scale factor of 8.
+    """
+
+    latents: LatentsField = InputField(
+        description=FieldDescriptions.latents,
+        input=Input.Connection,
+    )
+    x: int = InputField(
+        ge=0,
+        multiple_of=LATENT_SCALE_FACTOR,
+        description="The left x coordinate (in px) of the crop rectangle in image space. This value will be converted to a dimension in latent space.",
+    )
+    y: int = InputField(
+        ge=0,
+        multiple_of=LATENT_SCALE_FACTOR,
+        description="The top y coordinate (in px) of the crop rectangle in image space. This value will be converted to a dimension in latent space.",
+    )
+    width: int = InputField(
+        ge=1,
+        multiple_of=LATENT_SCALE_FACTOR,
+        description="The width (in px) of the crop rectangle in image space. This value will be converted to a dimension in latent space.",
+    )
+    height: int = InputField(
+        ge=1,
+        multiple_of=LATENT_SCALE_FACTOR,
+        description="The height (in px) of the crop rectangle in image space. This value will be converted to a dimension in latent space.",
+    )
+
+    def invoke(self, context: InvocationContext) -> LatentsOutput:
+        latents = context.tensors.load(self.latents.latents_name)
+
+        x1 = self.x // LATENT_SCALE_FACTOR
+        y1 = self.y // LATENT_SCALE_FACTOR
+        x2 = x1 + (self.width // LATENT_SCALE_FACTOR)
+        y2 = y1 + (self.height // LATENT_SCALE_FACTOR)
+
+        cropped_latents = latents[..., y1:y2, x1:x2]
+
+        name = context.tensors.save(tensor=cropped_latents)
+
+        return LatentsOutput.build(latents_name=name, latents=cropped_latents)

invokeai/app/invocations/denoise_latents.py

@@ -0,0 +1,812 @@
+# Copyright (c) 2023 Kyle Schouviller (https://github.com/kyle0654)
+import copy
+import inspect
+from contextlib import ExitStack
+from typing import Any, Dict, Iterator, List, Optional, Tuple, Union
+
+import torch
+import torchvision
+import torchvision.transforms as T
+from diffusers.configuration_utils import ConfigMixin
+from diffusers.models.adapter import T2IAdapter
+from diffusers.models.unets.unet_2d_condition import UNet2DConditionModel
+from diffusers.schedulers.scheduling_dpmsolver_sde import DPMSolverSDEScheduler
+from diffusers.schedulers.scheduling_tcd import TCDScheduler
+from diffusers.schedulers.scheduling_utils import SchedulerMixin as Scheduler
+from pydantic import field_validator
+from torchvision.transforms.functional import resize as tv_resize
+from transformers import CLIPVisionModelWithProjection
+
+from invokeai.app.invocations.baseinvocation import BaseInvocation, invocation
+from invokeai.app.invocations.constants import LATENT_SCALE_FACTOR, SCHEDULER_NAME_VALUES
+from invokeai.app.invocations.controlnet_image_processors import ControlField
+from invokeai.app.invocations.fields import (
+    ConditioningField,
+    DenoiseMaskField,
+    FieldDescriptions,
+    Input,
+    InputField,
+    LatentsField,
+    UIType,
+)
+from invokeai.app.invocations.ip_adapter import IPAdapterField
+from invokeai.app.invocations.model import ModelIdentifierField, UNetField
+from invokeai.app.invocations.primitives import LatentsOutput
+from invokeai.app.invocations.t2i_adapter import T2IAdapterField
+from invokeai.app.services.shared.invocation_context import InvocationContext
+from invokeai.app.util.controlnet_utils import prepare_control_image
+from invokeai.backend.ip_adapter.ip_adapter import IPAdapter
+from invokeai.backend.lora import LoRAModelRaw
+from invokeai.backend.model_manager import BaseModelType
+from invokeai.backend.model_patcher import ModelPatcher
+from invokeai.backend.stable_diffusion import PipelineIntermediateState, set_seamless
+from invokeai.backend.stable_diffusion.diffusers_pipeline import (
+    ControlNetData,
+    StableDiffusionGeneratorPipeline,
+    T2IAdapterData,
+)
+from invokeai.backend.stable_diffusion.diffusion.conditioning_data import (
+    BasicConditioningInfo,
+    IPAdapterConditioningInfo,
+    IPAdapterData,
+    Range,
+    SDXLConditioningInfo,
+    TextConditioningData,
+    TextConditioningRegions,
+)
+from invokeai.backend.stable_diffusion.schedulers import SCHEDULER_MAP
+from invokeai.backend.util.devices import TorchDevice
+from invokeai.backend.util.mask import to_standard_float_mask
+from invokeai.backend.util.silence_warnings import SilenceWarnings
+
+
+def get_scheduler(
+    context: InvocationContext,
+    scheduler_info: ModelIdentifierField,
+    scheduler_name: str,
+    seed: int,
+) -> Scheduler:
+    scheduler_class, scheduler_extra_config = SCHEDULER_MAP.get(scheduler_name, SCHEDULER_MAP["ddim"])
+    orig_scheduler_info = context.models.load(scheduler_info)
+    with orig_scheduler_info as orig_scheduler:
+        scheduler_config = orig_scheduler.config
+
+    if "_backup" in scheduler_config:
+        scheduler_config = scheduler_config["_backup"]
+    scheduler_config = {
+        **scheduler_config,
+        **scheduler_extra_config,  # FIXME
+        "_backup": scheduler_config,
+    }
+
+    # make dpmpp_sde reproducable(seed can be passed only in initializer)
+    if scheduler_class is DPMSolverSDEScheduler:
+        scheduler_config["noise_sampler_seed"] = seed
+
+    scheduler = scheduler_class.from_config(scheduler_config)
+
+    # hack copied over from generate.py
+    if not hasattr(scheduler, "uses_inpainting_model"):
+        scheduler.uses_inpainting_model = lambda: False
+    assert isinstance(scheduler, Scheduler)
+    return scheduler
+
+
+@invocation(
+    "denoise_latents",
+    title="Denoise Latents",
+    tags=["latents", "denoise", "txt2img", "t2i", "t2l", "img2img", "i2i", "l2l"],
+    category="latents",
+    version="1.5.3",
+)
+class DenoiseLatentsInvocation(BaseInvocation):
+    """Denoises noisy latents to decodable images"""
+
+    positive_conditioning: Union[ConditioningField, list[ConditioningField]] = InputField(
+        description=FieldDescriptions.positive_cond, input=Input.Connection, ui_order=0
+    )
+    negative_conditioning: Union[ConditioningField, list[ConditioningField]] = InputField(
+        description=FieldDescriptions.negative_cond, input=Input.Connection, ui_order=1
+    )
+    noise: Optional[LatentsField] = InputField(
+        default=None,
+        description=FieldDescriptions.noise,
+        input=Input.Connection,
+        ui_order=3,
+    )
+    steps: int = InputField(default=10, gt=0, description=FieldDescriptions.steps)
+    cfg_scale: Union[float, List[float]] = InputField(
+        default=7.5, description=FieldDescriptions.cfg_scale, title="CFG Scale"
+    )
+    denoising_start: float = InputField(
+        default=0.0,
+        ge=0,
+        le=1,
+        description=FieldDescriptions.denoising_start,
+    )
+    denoising_end: float = InputField(default=1.0, ge=0, le=1, description=FieldDescriptions.denoising_end)
+    scheduler: SCHEDULER_NAME_VALUES = InputField(
+        default="euler",
+        description=FieldDescriptions.scheduler,
+        ui_type=UIType.Scheduler,
+    )
+    unet: UNetField = InputField(
+        description=FieldDescriptions.unet,
+        input=Input.Connection,
+        title="UNet",
+        ui_order=2,
+    )
+    control: Optional[Union[ControlField, list[ControlField]]] = InputField(
+        default=None,
+        input=Input.Connection,
+        ui_order=5,
+    )
+    ip_adapter: Optional[Union[IPAdapterField, list[IPAdapterField]]] = InputField(
+        description=FieldDescriptions.ip_adapter,
+        title="IP-Adapter",
+        default=None,
+        input=Input.Connection,
+        ui_order=6,
+    )
+    t2i_adapter: Optional[Union[T2IAdapterField, list[T2IAdapterField]]] = InputField(
+        description=FieldDescriptions.t2i_adapter,
+        title="T2I-Adapter",
+        default=None,
+        input=Input.Connection,
+        ui_order=7,
+    )
+    cfg_rescale_multiplier: float = InputField(
+        title="CFG Rescale Multiplier", default=0, ge=0, lt=1, description=FieldDescriptions.cfg_rescale_multiplier
+    )
+    latents: Optional[LatentsField] = InputField(
+        default=None,
+        description=FieldDescriptions.latents,
+        input=Input.Connection,
+        ui_order=4,
+    )
+    denoise_mask: Optional[DenoiseMaskField] = InputField(
+        default=None,
+        description=FieldDescriptions.mask,
+        input=Input.Connection,
+        ui_order=8,
+    )
+
+    @field_validator("cfg_scale")
+    def ge_one(cls, v: Union[List[float], float]) -> Union[List[float], float]:
+        """validate that all cfg_scale values are >= 1"""
+        if isinstance(v, list):
+            for i in v:
+                if i < 1:
+                    raise ValueError("cfg_scale must be greater than 1")
+        else:
+            if v < 1:
+                raise ValueError("cfg_scale must be greater than 1")
+        return v
+
+    def _get_text_embeddings_and_masks(
+        self,
+        cond_list: list[ConditioningField],
+        context: InvocationContext,
+        device: torch.device,
+        dtype: torch.dtype,
+    ) -> tuple[Union[list[BasicConditioningInfo], list[SDXLConditioningInfo]], list[Optional[torch.Tensor]]]:
+        """Get the text embeddings and masks from the input conditioning fields."""
+        text_embeddings: Union[list[BasicConditioningInfo], list[SDXLConditioningInfo]] = []
+        text_embeddings_masks: list[Optional[torch.Tensor]] = []
+        for cond in cond_list:
+            cond_data = copy.deepcopy(context.conditioning.load(cond.conditioning_name))
+            text_embeddings.append(cond_data.conditionings[0].to(device=device, dtype=dtype))
+            mask = cond.mask
+            if mask is not None:
+                mask = context.tensors.load(mask.tensor_name)
+            text_embeddings_masks.append(mask)
+
+        return text_embeddings, text_embeddings_masks
+
+    def _preprocess_regional_prompt_mask(
+        self, mask: Optional[torch.Tensor], target_height: int, target_width: int, dtype: torch.dtype
+    ) -> torch.Tensor:
+        """Preprocess a regional prompt mask to match the target height and width.
+        If mask is None, returns a mask of all ones with the target height and width.
+        If mask is not None, resizes the mask to the target height and width using 'nearest' interpolation.
+
+        Returns:
+            torch.Tensor: The processed mask. shape: (1, 1, target_height, target_width).
+        """
+
+        if mask is None:
+            return torch.ones((1, 1, target_height, target_width), dtype=dtype)
+
+        mask = to_standard_float_mask(mask, out_dtype=dtype)
+
+        tf = torchvision.transforms.Resize(
+            (target_height, target_width), interpolation=torchvision.transforms.InterpolationMode.NEAREST
+        )
+
+        # Add a batch dimension to the mask, because torchvision expects shape (batch, channels, h, w).
+        mask = mask.unsqueeze(0)  # Shape: (1, h, w) -> (1, 1, h, w)
+        resized_mask = tf(mask)
+        assert isinstance(resized_mask, torch.Tensor)
+        return resized_mask
+
+    def _concat_regional_text_embeddings(
+        self,
+        text_conditionings: Union[list[BasicConditioningInfo], list[SDXLConditioningInfo]],
+        masks: Optional[list[Optional[torch.Tensor]]],
+        latent_height: int,
+        latent_width: int,
+        dtype: torch.dtype,
+    ) -> tuple[Union[BasicConditioningInfo, SDXLConditioningInfo], Optional[TextConditioningRegions]]:
+        """Concatenate regional text embeddings into a single embedding and track the region masks accordingly."""
+        if masks is None:
+            masks = [None] * len(text_conditionings)
+        assert len(text_conditionings) == len(masks)
+
+        is_sdxl = type(text_conditionings[0]) is SDXLConditioningInfo
+
+        all_masks_are_none = all(mask is None for mask in masks)
+
+        text_embedding = []
+        pooled_embedding = None
+        add_time_ids = None
+        cur_text_embedding_len = 0
+        processed_masks = []
+        embedding_ranges = []
+
+        for prompt_idx, text_embedding_info in enumerate(text_conditionings):
+            mask = masks[prompt_idx]
+
+            if is_sdxl:
+                # We choose a random SDXLConditioningInfo's pooled_embeds and add_time_ids here, with a preference for
+                # prompts without a mask. We prefer prompts without a mask, because they are more likely to contain
+                # global prompt information.  In an ideal case, there should be exactly one global prompt without a
+                # mask, but we don't enforce this.
+
+                # HACK(ryand): The fact that we have to choose a single pooled_embedding and add_time_ids here is a
+                # fundamental interface issue. The SDXL Compel nodes are not designed to be used in the way that we use
+                # them for regional prompting. Ideally, the DenoiseLatents invocation should accept a single
+                # pooled_embeds tensor and a list of standard text embeds with region masks. This change would be a
+                # pretty major breaking change to a popular node, so for now we use this hack.
+                if pooled_embedding is None or mask is None:
+                    pooled_embedding = text_embedding_info.pooled_embeds
+                if add_time_ids is None or mask is None:
+                    add_time_ids = text_embedding_info.add_time_ids
+
+            text_embedding.append(text_embedding_info.embeds)
+            if not all_masks_are_none:
+                embedding_ranges.append(
+                    Range(
+                        start=cur_text_embedding_len, end=cur_text_embedding_len + text_embedding_info.embeds.shape[1]
+                    )
+                )
+                processed_masks.append(
+                    self._preprocess_regional_prompt_mask(mask, latent_height, latent_width, dtype=dtype)
+                )
+
+            cur_text_embedding_len += text_embedding_info.embeds.shape[1]
+
+        text_embedding = torch.cat(text_embedding, dim=1)
+        assert len(text_embedding.shape) == 3  # batch_size, seq_len, token_len
+
+        regions = None
+        if not all_masks_are_none:
+            regions = TextConditioningRegions(
+                masks=torch.cat(processed_masks, dim=1),
+                ranges=embedding_ranges,
+            )
+
+        if is_sdxl:
+            return (
+                SDXLConditioningInfo(embeds=text_embedding, pooled_embeds=pooled_embedding, add_time_ids=add_time_ids),
+                regions,
+            )
+        return BasicConditioningInfo(embeds=text_embedding), regions
+
+    def get_conditioning_data(
+        self,
+        context: InvocationContext,
+        unet: UNet2DConditionModel,
+        latent_height: int,
+        latent_width: int,
+    ) -> TextConditioningData:
+        # Normalize self.positive_conditioning and self.negative_conditioning to lists.
+        cond_list = self.positive_conditioning
+        if not isinstance(cond_list, list):
+            cond_list = [cond_list]
+        uncond_list = self.negative_conditioning
+        if not isinstance(uncond_list, list):
+            uncond_list = [uncond_list]
+
+        cond_text_embeddings, cond_text_embedding_masks = self._get_text_embeddings_and_masks(
+            cond_list, context, unet.device, unet.dtype
+        )
+        uncond_text_embeddings, uncond_text_embedding_masks = self._get_text_embeddings_and_masks(
+            uncond_list, context, unet.device, unet.dtype
+        )
+
+        cond_text_embedding, cond_regions = self._concat_regional_text_embeddings(
+            text_conditionings=cond_text_embeddings,
+            masks=cond_text_embedding_masks,
+            latent_height=latent_height,
+            latent_width=latent_width,
+            dtype=unet.dtype,
+        )
+        uncond_text_embedding, uncond_regions = self._concat_regional_text_embeddings(
+            text_conditionings=uncond_text_embeddings,
+            masks=uncond_text_embedding_masks,
+            latent_height=latent_height,
+            latent_width=latent_width,
+            dtype=unet.dtype,
+        )
+
+        if isinstance(self.cfg_scale, list):
+            assert (
+                len(self.cfg_scale) == self.steps
+            ), "cfg_scale (list) must have the same length as the number of steps"
+
+        conditioning_data = TextConditioningData(
+            uncond_text=uncond_text_embedding,
+            cond_text=cond_text_embedding,
+            uncond_regions=uncond_regions,
+            cond_regions=cond_regions,
+            guidance_scale=self.cfg_scale,
+            guidance_rescale_multiplier=self.cfg_rescale_multiplier,
+        )
+        return conditioning_data
+
+    def create_pipeline(
+        self,
+        unet: UNet2DConditionModel,
+        scheduler: Scheduler,
+    ) -> StableDiffusionGeneratorPipeline:
+        class FakeVae:
+            class FakeVaeConfig:
+                def __init__(self) -> None:
+                    self.block_out_channels = [0]
+
+            def __init__(self) -> None:
+                self.config = FakeVae.FakeVaeConfig()
+
+        return StableDiffusionGeneratorPipeline(
+            vae=FakeVae(),  # TODO: oh...
+            text_encoder=None,
+            tokenizer=None,
+            unet=unet,
+            scheduler=scheduler,
+            safety_checker=None,
+            feature_extractor=None,
+            requires_safety_checker=False,
+        )
+
+    def prep_control_data(
+        self,
+        context: InvocationContext,
+        control_input: Optional[Union[ControlField, List[ControlField]]],
+        latents_shape: List[int],
+        exit_stack: ExitStack,
+        do_classifier_free_guidance: bool = True,
+    ) -> Optional[List[ControlNetData]]:
+        # Assuming fixed dimensional scaling of LATENT_SCALE_FACTOR.
+        control_height_resize = latents_shape[2] * LATENT_SCALE_FACTOR
+        control_width_resize = latents_shape[3] * LATENT_SCALE_FACTOR
+        if control_input is None:
+            control_list = None
+        elif isinstance(control_input, list) and len(control_input) == 0:
+            control_list = None
+        elif isinstance(control_input, ControlField):
+            control_list = [control_input]
+        elif isinstance(control_input, list) and len(control_input) > 0 and isinstance(control_input[0], ControlField):
+            control_list = control_input
+        else:
+            control_list = None
+        if control_list is None:
+            return None
+        # After above handling, any control that is not None should now be of type list[ControlField].
+
+        # FIXME: add checks to skip entry if model or image is None
+        #        and if weight is None, populate with default 1.0?
+        controlnet_data = []
+        for control_info in control_list:
+            control_model = exit_stack.enter_context(context.models.load(control_info.control_model))
+
+            # control_models.append(control_model)
+            control_image_field = control_info.image
+            input_image = context.images.get_pil(control_image_field.image_name)
+            # self.image.image_type, self.image.image_name
+            # FIXME: still need to test with different widths, heights, devices, dtypes
+            #        and add in batch_size, num_images_per_prompt?
+            #        and do real check for classifier_free_guidance?
+            # prepare_control_image should return torch.Tensor of shape(batch_size, 3, height, width)
+            control_image = prepare_control_image(
+                image=input_image,
+                do_classifier_free_guidance=do_classifier_free_guidance,
+                width=control_width_resize,
+                height=control_height_resize,
+                # batch_size=batch_size * num_images_per_prompt,
+                # num_images_per_prompt=num_images_per_prompt,
+                device=control_model.device,
+                dtype=control_model.dtype,
+                control_mode=control_info.control_mode,
+                resize_mode=control_info.resize_mode,
+            )
+            control_item = ControlNetData(
+                model=control_model,  # model object
+                image_tensor=control_image,
+                weight=control_info.control_weight,
+                begin_step_percent=control_info.begin_step_percent,
+                end_step_percent=control_info.end_step_percent,
+                control_mode=control_info.control_mode,
+                # any resizing needed should currently be happening in prepare_control_image(),
+                #    but adding resize_mode to ControlNetData in case needed in the future
+                resize_mode=control_info.resize_mode,
+            )
+            controlnet_data.append(control_item)
+            # MultiControlNetModel has been refactored out, just need list[ControlNetData]
+
+        return controlnet_data
+
+    def prep_ip_adapter_image_prompts(
+        self,
+        context: InvocationContext,
+        ip_adapters: List[IPAdapterField],
+    ) -> List[Tuple[torch.Tensor, torch.Tensor]]:
+        """Run the IPAdapter CLIPVisionModel, returning image prompt embeddings."""
+        image_prompts = []
+        for single_ip_adapter in ip_adapters:
+            with context.models.load(single_ip_adapter.ip_adapter_model) as ip_adapter_model:
+                assert isinstance(ip_adapter_model, IPAdapter)
+                image_encoder_model_info = context.models.load(single_ip_adapter.image_encoder_model)
+                # `single_ip_adapter.image` could be a list or a single ImageField. Normalize to a list here.
+                single_ipa_image_fields = single_ip_adapter.image
+                if not isinstance(single_ipa_image_fields, list):
+                    single_ipa_image_fields = [single_ipa_image_fields]
+
+                single_ipa_images = [context.images.get_pil(image.image_name) for image in single_ipa_image_fields]
+                with image_encoder_model_info as image_encoder_model:
+                    assert isinstance(image_encoder_model, CLIPVisionModelWithProjection)
+                    # Get image embeddings from CLIP and ImageProjModel.
+                    image_prompt_embeds, uncond_image_prompt_embeds = ip_adapter_model.get_image_embeds(
+                        single_ipa_images, image_encoder_model
+                    )
+                    image_prompts.append((image_prompt_embeds, uncond_image_prompt_embeds))
+
+        return image_prompts
+
+    def prep_ip_adapter_data(
+        self,
+        context: InvocationContext,
+        ip_adapters: List[IPAdapterField],
+        image_prompts: List[Tuple[torch.Tensor, torch.Tensor]],
+        exit_stack: ExitStack,
+        latent_height: int,
+        latent_width: int,
+        dtype: torch.dtype,
+    ) -> Optional[List[IPAdapterData]]:
+        """If IP-Adapter is enabled, then this function loads the requisite models and adds the image prompt conditioning data."""
+        ip_adapter_data_list = []
+        for single_ip_adapter, (image_prompt_embeds, uncond_image_prompt_embeds) in zip(
+            ip_adapters, image_prompts, strict=True
+        ):
+            ip_adapter_model = exit_stack.enter_context(context.models.load(single_ip_adapter.ip_adapter_model))
+
+            mask_field = single_ip_adapter.mask
+            mask = context.tensors.load(mask_field.tensor_name) if mask_field is not None else None
+            mask = self._preprocess_regional_prompt_mask(mask, latent_height, latent_width, dtype=dtype)
+
+            ip_adapter_data_list.append(
+                IPAdapterData(
+                    ip_adapter_model=ip_adapter_model,
+                    weight=single_ip_adapter.weight,
+                    target_blocks=single_ip_adapter.target_blocks,
+                    begin_step_percent=single_ip_adapter.begin_step_percent,
+                    end_step_percent=single_ip_adapter.end_step_percent,
+                    ip_adapter_conditioning=IPAdapterConditioningInfo(image_prompt_embeds, uncond_image_prompt_embeds),
+                    mask=mask,
+                )
+            )
+
+        return ip_adapter_data_list if len(ip_adapter_data_list) > 0 else None
+
+    def run_t2i_adapters(
+        self,
+        context: InvocationContext,
+        t2i_adapter: Optional[Union[T2IAdapterField, list[T2IAdapterField]]],
+        latents_shape: list[int],
+        do_classifier_free_guidance: bool,
+    ) -> Optional[list[T2IAdapterData]]:
+        if t2i_adapter is None:
+            return None
+
+        # Handle the possibility that t2i_adapter could be a list or a single T2IAdapterField.
+        if isinstance(t2i_adapter, T2IAdapterField):
+            t2i_adapter = [t2i_adapter]
+
+        if len(t2i_adapter) == 0:
+            return None
+
+        t2i_adapter_data = []
+        for t2i_adapter_field in t2i_adapter:
+            t2i_adapter_model_config = context.models.get_config(t2i_adapter_field.t2i_adapter_model.key)
+            t2i_adapter_loaded_model = context.models.load(t2i_adapter_field.t2i_adapter_model)
+            image = context.images.get_pil(t2i_adapter_field.image.image_name)
+
+            # The max_unet_downscale is the maximum amount that the UNet model downscales the latent image internally.
+            if t2i_adapter_model_config.base == BaseModelType.StableDiffusion1:
+                max_unet_downscale = 8
+            elif t2i_adapter_model_config.base == BaseModelType.StableDiffusionXL:
+                max_unet_downscale = 4
+            else:
+                raise ValueError(f"Unexpected T2I-Adapter base model type: '{t2i_adapter_model_config.base}'.")
+
+            t2i_adapter_model: T2IAdapter
+            with t2i_adapter_loaded_model as t2i_adapter_model:
+                total_downscale_factor = t2i_adapter_model.total_downscale_factor
+
+                # Resize the T2I-Adapter input image.
+                # We select the resize dimensions so that after the T2I-Adapter's total_downscale_factor is applied, the
+                # result will match the latent image's dimensions after max_unet_downscale is applied.
+                t2i_input_height = latents_shape[2] // max_unet_downscale * total_downscale_factor
+                t2i_input_width = latents_shape[3] // max_unet_downscale * total_downscale_factor
+
+                # Note: We have hard-coded `do_classifier_free_guidance=False`. This is because we only want to prepare
+                # a single image. If CFG is enabled, we will duplicate the resultant tensor after applying the
+                # T2I-Adapter model.
+                #
+                # Note: We re-use the `prepare_control_image(...)` from ControlNet for T2I-Adapter, because it has many
+                # of the same requirements (e.g. preserving binary masks during resize).
+                t2i_image = prepare_control_image(
+                    image=image,
+                    do_classifier_free_guidance=False,
+                    width=t2i_input_width,
+                    height=t2i_input_height,
+                    num_channels=t2i_adapter_model.config["in_channels"],  # mypy treats this as a FrozenDict
+                    device=t2i_adapter_model.device,
+                    dtype=t2i_adapter_model.dtype,
+                    resize_mode=t2i_adapter_field.resize_mode,
+                )
+
+                adapter_state = t2i_adapter_model(t2i_image)
+
+            if do_classifier_free_guidance:
+                for idx, value in enumerate(adapter_state):
+                    adapter_state[idx] = torch.cat([value] * 2, dim=0)
+
+            t2i_adapter_data.append(
+                T2IAdapterData(
+                    adapter_state=adapter_state,
+                    weight=t2i_adapter_field.weight,
+                    begin_step_percent=t2i_adapter_field.begin_step_percent,
+                    end_step_percent=t2i_adapter_field.end_step_percent,
+                )
+            )
+
+        return t2i_adapter_data
+
+    # original idea by https://github.com/AmericanPresidentJimmyCarter
+    # TODO: research more for second order schedulers timesteps
+    def init_scheduler(
+        self,
+        scheduler: Union[Scheduler, ConfigMixin],
+        device: torch.device,
+        steps: int,
+        denoising_start: float,
+        denoising_end: float,
+        seed: int,
+    ) -> Tuple[int, List[int], int, Dict[str, Any]]:
+        assert isinstance(scheduler, ConfigMixin)
+        if scheduler.config.get("cpu_only", False):
+            scheduler.set_timesteps(steps, device="cpu")
+            timesteps = scheduler.timesteps.to(device=device)
+        else:
+            scheduler.set_timesteps(steps, device=device)
+            timesteps = scheduler.timesteps
+
+        # skip greater order timesteps
+        _timesteps = timesteps[:: scheduler.order]
+
+        # get start timestep index
+        t_start_val = int(round(scheduler.config["num_train_timesteps"] * (1 - denoising_start)))
+        t_start_idx = len(list(filter(lambda ts: ts >= t_start_val, _timesteps)))
+
+        # get end timestep index
+        t_end_val = int(round(scheduler.config["num_train_timesteps"] * (1 - denoising_end)))
+        t_end_idx = len(list(filter(lambda ts: ts >= t_end_val, _timesteps[t_start_idx:])))
+
+        # apply order to indexes
+        t_start_idx *= scheduler.order
+        t_end_idx *= scheduler.order
+
+        init_timestep = timesteps[t_start_idx : t_start_idx + 1]
+        timesteps = timesteps[t_start_idx : t_start_idx + t_end_idx]
+        num_inference_steps = len(timesteps) // scheduler.order
+
+        scheduler_step_kwargs: Dict[str, Any] = {}
+        scheduler_step_signature = inspect.signature(scheduler.step)
+        if "generator" in scheduler_step_signature.parameters:
+            # At some point, someone decided that schedulers that accept a generator should use the original seed with
+            # all bits flipped. I don't know the original rationale for this, but now we must keep it like this for
+            # reproducibility.
+            #
+            # These Invoke-supported schedulers accept a generator as of 2024-06-04:
+            #   - DDIMScheduler
+            #   - DDPMScheduler
+            #   - DPMSolverMultistepScheduler
+            #   - EulerAncestralDiscreteScheduler
+            #   - EulerDiscreteScheduler
+            #   - KDPM2AncestralDiscreteScheduler
+            #   - LCMScheduler
+            #   - TCDScheduler
+            scheduler_step_kwargs.update({"generator": torch.Generator(device=device).manual_seed(seed ^ 0xFFFFFFFF)})
+        if isinstance(scheduler, TCDScheduler):
+            scheduler_step_kwargs.update({"eta": 1.0})
+
+        return num_inference_steps, timesteps, init_timestep, scheduler_step_kwargs
+
+    def prep_inpaint_mask(
+        self, context: InvocationContext, latents: torch.Tensor
+    ) -> Tuple[Optional[torch.Tensor], Optional[torch.Tensor], bool]:
+        if self.denoise_mask is None:
+            return None, None, False
+
+        mask = context.tensors.load(self.denoise_mask.mask_name)
+        mask = tv_resize(mask, latents.shape[-2:], T.InterpolationMode.BILINEAR, antialias=False)
+        if self.denoise_mask.masked_latents_name is not None:
+            masked_latents = context.tensors.load(self.denoise_mask.masked_latents_name)
+        else:
+            masked_latents = torch.where(mask < 0.5, 0.0, latents)
+
+        return 1 - mask, masked_latents, self.denoise_mask.gradient
+
+    @torch.no_grad()
+    @SilenceWarnings()  # This quenches the NSFW nag from diffusers.
+    def invoke(self, context: InvocationContext) -> LatentsOutput:
+        seed = None
+        noise = None
+        if self.noise is not None:
+            noise = context.tensors.load(self.noise.latents_name)
+            seed = self.noise.seed
+
+        if self.latents is not None:
+            latents = context.tensors.load(self.latents.latents_name)
+            if seed is None:
+                seed = self.latents.seed
+
+            if noise is not None and noise.shape[1:] != latents.shape[1:]:
+                raise Exception(f"Incompatable 'noise' and 'latents' shapes: {latents.shape=} {noise.shape=}")
+
+        elif noise is not None:
+            latents = torch.zeros_like(noise)
+        else:
+            raise Exception("'latents' or 'noise' must be provided!")
+
+        if seed is None:
+            seed = 0
+
+        mask, masked_latents, gradient_mask = self.prep_inpaint_mask(context, latents)
+
+        # TODO(ryand): I have hard-coded `do_classifier_free_guidance=True` to mirror the behaviour of ControlNets,
+        # below. Investigate whether this is appropriate.
+        t2i_adapter_data = self.run_t2i_adapters(
+            context,
+            self.t2i_adapter,
+            latents.shape,
+            do_classifier_free_guidance=True,
+        )
+
+        ip_adapters: List[IPAdapterField] = []
+        if self.ip_adapter is not None:
+            # ip_adapter could be a list or a single IPAdapterField. Normalize to a list here.
+            if isinstance(self.ip_adapter, list):
+                ip_adapters = self.ip_adapter
+            else:
+                ip_adapters = [self.ip_adapter]
+
+        # If there are IP adapters, the following line runs the adapters' CLIPVision image encoders to return
+        # a series of image conditioning embeddings. This is being done here rather than in the
+        # big model context below in order to use less VRAM on low-VRAM systems.
+        # The image prompts are then passed to prep_ip_adapter_data().
+        image_prompts = self.prep_ip_adapter_image_prompts(context=context, ip_adapters=ip_adapters)
+
+        # get the unet's config so that we can pass the base to dispatch_progress()
+        unet_config = context.models.get_config(self.unet.unet.key)
+
+        def step_callback(state: PipelineIntermediateState) -> None:
+            context.util.sd_step_callback(state, unet_config.base)
+
+        def _lora_loader() -> Iterator[Tuple[LoRAModelRaw, float]]:
+            for lora in self.unet.loras:
+                lora_info = context.models.load(lora.lora)
+                assert isinstance(lora_info.model, LoRAModelRaw)
+                yield (lora_info.model, lora.weight)
+                del lora_info
+            return
+
+        unet_info = context.models.load(self.unet.unet)
+        assert isinstance(unet_info.model, UNet2DConditionModel)
+        with (
+            ExitStack() as exit_stack,
+            unet_info.model_on_device() as (model_state_dict, unet),
+            ModelPatcher.apply_freeu(unet, self.unet.freeu_config),
+            set_seamless(unet, self.unet.seamless_axes),  # FIXME
+            # Apply the LoRA after unet has been moved to its target device for faster patching.
+            ModelPatcher.apply_lora_unet(
+                unet,
+                loras=_lora_loader(),
+                model_state_dict=model_state_dict,
+            ),
+        ):
+            assert isinstance(unet, UNet2DConditionModel)
+            latents = latents.to(device=unet.device, dtype=unet.dtype)
+            if noise is not None:
+                noise = noise.to(device=unet.device, dtype=unet.dtype)
+            if mask is not None:
+                mask = mask.to(device=unet.device, dtype=unet.dtype)
+            if masked_latents is not None:
+                masked_latents = masked_latents.to(device=unet.device, dtype=unet.dtype)
+
+            scheduler = get_scheduler(
+                context=context,
+                scheduler_info=self.unet.scheduler,
+                scheduler_name=self.scheduler,
+                seed=seed,
+            )
+
+            pipeline = self.create_pipeline(unet, scheduler)
+
+            _, _, latent_height, latent_width = latents.shape
+            conditioning_data = self.get_conditioning_data(
+                context=context, unet=unet, latent_height=latent_height, latent_width=latent_width
+            )
+
+            controlnet_data = self.prep_control_data(
+                context=context,
+                control_input=self.control,
+                latents_shape=latents.shape,
+                # do_classifier_free_guidance=(self.cfg_scale >= 1.0))
+                do_classifier_free_guidance=True,
+                exit_stack=exit_stack,
+            )
+
+            ip_adapter_data = self.prep_ip_adapter_data(
+                context=context,
+                ip_adapters=ip_adapters,
+                image_prompts=image_prompts,
+                exit_stack=exit_stack,
+                latent_height=latent_height,
+                latent_width=latent_width,
+                dtype=unet.dtype,
+            )
+
+            num_inference_steps, timesteps, init_timestep, scheduler_step_kwargs = self.init_scheduler(
+                scheduler,
+                device=unet.device,
+                steps=self.steps,
+                denoising_start=self.denoising_start,
+                denoising_end=self.denoising_end,
+                seed=seed,
+            )
+
+            result_latents = pipeline.latents_from_embeddings(
+                latents=latents,
+                timesteps=timesteps,
+                init_timestep=init_timestep,
+                noise=noise,
+                seed=seed,
+                mask=mask,
+                masked_latents=masked_latents,
+                gradient_mask=gradient_mask,
+                num_inference_steps=num_inference_steps,
+                scheduler_step_kwargs=scheduler_step_kwargs,
+                conditioning_data=conditioning_data,
+                control_data=controlnet_data,
+                ip_adapter_data=ip_adapter_data,
+                t2i_adapter_data=t2i_adapter_data,
+                callback=step_callback,
+            )
+
+        # https://discuss.huggingface.co/t/memory-usage-by-later-pipeline-stages/23699
+        result_latents = result_latents.to("cpu")
+        TorchDevice.empty_cache()
+
+        name = context.tensors.save(tensor=result_latents)
+        return LatentsOutput.build(latents_name=name, latents=result_latents, seed=None)

invokeai/app/invocations/ideal_size.py

@@ -0,0 +1,65 @@
+import math
+from typing import Tuple
+
+from invokeai.app.invocations.baseinvocation import BaseInvocation, BaseInvocationOutput, invocation, invocation_output
+from invokeai.app.invocations.constants import LATENT_SCALE_FACTOR
+from invokeai.app.invocations.fields import FieldDescriptions, InputField, OutputField
+from invokeai.app.invocations.model import UNetField
+from invokeai.app.services.shared.invocation_context import InvocationContext
+from invokeai.backend.model_manager.config import BaseModelType
+
+
+@invocation_output("ideal_size_output")
+class IdealSizeOutput(BaseInvocationOutput):
+    """Base class for invocations that output an image"""
+
+    width: int = OutputField(description="The ideal width of the image (in pixels)")
+    height: int = OutputField(description="The ideal height of the image (in pixels)")
+
+
+@invocation(
+    "ideal_size",
+    title="Ideal Size",
+    tags=["latents", "math", "ideal_size"],
+    version="1.0.3",
+)
+class IdealSizeInvocation(BaseInvocation):
+    """Calculates the ideal size for generation to avoid duplication"""
+
+    width: int = InputField(default=1024, description="Final image width")
+    height: int = InputField(default=576, description="Final image height")
+    unet: UNetField = InputField(default=None, description=FieldDescriptions.unet)
+    multiplier: float = InputField(
+        default=1.0,
+        description="Amount to multiply the model's dimensions by when calculating the ideal size (may result in "
+        "initial generation artifacts if too large)",
+    )
+
+    def trim_to_multiple_of(self, *args: int, multiple_of: int = LATENT_SCALE_FACTOR) -> Tuple[int, ...]:
+        return tuple((x - x % multiple_of) for x in args)
+
+    def invoke(self, context: InvocationContext) -> IdealSizeOutput:
+        unet_config = context.models.get_config(self.unet.unet.key)
+        aspect = self.width / self.height
+        dimension: float = 512
+        if unet_config.base == BaseModelType.StableDiffusion2:
+            dimension = 768
+        elif unet_config.base == BaseModelType.StableDiffusionXL:
+            dimension = 1024
+        dimension = dimension * self.multiplier
+        min_dimension = math.floor(dimension * 0.5)
+        model_area = dimension * dimension  # hardcoded for now since all models are trained on square images
+
+        if aspect > 1.0:
+            init_height = max(min_dimension, math.sqrt(model_area / aspect))
+            init_width = init_height * aspect
+        else:
+            init_width = max(min_dimension, math.sqrt(model_area * aspect))
+            init_height = init_width / aspect
+
+        scaled_width, scaled_height = self.trim_to_multiple_of(
+            math.floor(init_width),
+            math.floor(init_height),
+        )
+
+        return IdealSizeOutput(width=scaled_width, height=scaled_height)

invokeai/app/invocations/image.py

@@ -1,6 +1,5 @@
 # Copyright (c) 2022 Kyle Schouviller (https://github.com/kyle0654)
 
-from pathlib import Path
 from typing import Literal, Optional
 
 import cv2
@@ -504,7 +503,7 @@ class ImageInverseLerpInvocation(BaseInvocation, WithMetadata, WithBoard):
     title="Blur NSFW Image",
     tags=["image", "nsfw"],
     category="image",
-    version="1.2.2",
+    version="1.2.3",
 )
 class ImageNSFWBlurInvocation(BaseInvocation, WithMetadata, WithBoard):
     """Add blur to NSFW-flagged images"""
@@ -516,23 +515,12 @@ class ImageNSFWBlurInvocation(BaseInvocation, WithMetadata, WithBoard):
 
         logger = context.logger
         logger.debug("Running NSFW checker")
-        if SafetyChecker.has_nsfw_concept(image):
-            logger.info("A potentially NSFW image has been detected. Image will be blurred.")
-            blurry_image = image.filter(filter=ImageFilter.GaussianBlur(radius=32))
-            caution = self._get_caution_img()
-            blurry_image.paste(caution, (0, 0), caution)
-            image = blurry_image
+        image = SafetyChecker.blur_if_nsfw(image)
 
         image_dto = context.images.save(image=image)
 
         return ImageOutput.build(image_dto)
 
-    def _get_caution_img(self) -> Image.Image:
-        import invokeai.app.assets.images as image_assets
-
-        caution = Image.open(Path(image_assets.__path__[0]) / "caution.png")
-        return caution.resize((caution.width // 2, caution.height // 2))
-
 
 @invocation(
     "img_watermark",

invokeai/app/invocations/image_to_latents.py

@@ -0,0 +1,125 @@
+from functools import singledispatchmethod
+
+import einops
+import torch
+from diffusers.models.attention_processor import (
+    AttnProcessor2_0,
+    LoRAAttnProcessor2_0,
+    LoRAXFormersAttnProcessor,
+    XFormersAttnProcessor,
+)
+from diffusers.models.autoencoders.autoencoder_kl import AutoencoderKL
+from diffusers.models.autoencoders.autoencoder_tiny import AutoencoderTiny
+
+from invokeai.app.invocations.baseinvocation import BaseInvocation, invocation
+from invokeai.app.invocations.constants import DEFAULT_PRECISION
+from invokeai.app.invocations.fields import (
+    FieldDescriptions,
+    ImageField,
+    Input,
+    InputField,
+)
+from invokeai.app.invocations.model import VAEField
+from invokeai.app.invocations.primitives import LatentsOutput
+from invokeai.app.services.shared.invocation_context import InvocationContext
+from invokeai.backend.model_manager import LoadedModel
+from invokeai.backend.stable_diffusion.diffusers_pipeline import image_resized_to_grid_as_tensor
+
+
+@invocation(
+    "i2l",
+    title="Image to Latents",
+    tags=["latents", "image", "vae", "i2l"],
+    category="latents",
+    version="1.0.2",
+)
+class ImageToLatentsInvocation(BaseInvocation):
+    """Encodes an image into latents."""
+
+    image: ImageField = InputField(
+        description="The image to encode",
+    )
+    vae: VAEField = InputField(
+        description=FieldDescriptions.vae,
+        input=Input.Connection,
+    )
+    tiled: bool = InputField(default=False, description=FieldDescriptions.tiled)
+    fp32: bool = InputField(default=DEFAULT_PRECISION == torch.float32, description=FieldDescriptions.fp32)
+
+    @staticmethod
+    def vae_encode(vae_info: LoadedModel, upcast: bool, tiled: bool, image_tensor: torch.Tensor) -> torch.Tensor:
+        with vae_info as vae:
+            assert isinstance(vae, torch.nn.Module)
+            orig_dtype = vae.dtype
+            if upcast:
+                vae.to(dtype=torch.float32)
+
+                use_torch_2_0_or_xformers = hasattr(vae.decoder, "mid_block") and isinstance(
+                    vae.decoder.mid_block.attentions[0].processor,
+                    (
+                        AttnProcessor2_0,
+                        XFormersAttnProcessor,
+                        LoRAXFormersAttnProcessor,
+                        LoRAAttnProcessor2_0,
+                    ),
+                )
+                # if xformers or torch_2_0 is used attention block does not need
+                # to be in float32 which can save lots of memory
+                if use_torch_2_0_or_xformers:
+                    vae.post_quant_conv.to(orig_dtype)
+                    vae.decoder.conv_in.to(orig_dtype)
+                    vae.decoder.mid_block.to(orig_dtype)
+                # else:
+                #    latents = latents.float()
+
+            else:
+                vae.to(dtype=torch.float16)
+                # latents = latents.half()
+
+            if tiled:
+                vae.enable_tiling()
+            else:
+                vae.disable_tiling()
+
+            # non_noised_latents_from_image
+            image_tensor = image_tensor.to(device=vae.device, dtype=vae.dtype)
+            with torch.inference_mode():
+                latents = ImageToLatentsInvocation._encode_to_tensor(vae, image_tensor)
+
+            latents = vae.config.scaling_factor * latents
+            latents = latents.to(dtype=orig_dtype)
+
+        return latents
+
+    @torch.no_grad()
+    def invoke(self, context: InvocationContext) -> LatentsOutput:
+        image = context.images.get_pil(self.image.image_name)
+
+        vae_info = context.models.load(self.vae.vae)
+
+        image_tensor = image_resized_to_grid_as_tensor(image.convert("RGB"))
+        if image_tensor.dim() == 3:
+            image_tensor = einops.rearrange(image_tensor, "c h w -> 1 c h w")
+
+        latents = self.vae_encode(vae_info, self.fp32, self.tiled, image_tensor)
+
+        latents = latents.to("cpu")
+        name = context.tensors.save(tensor=latents)
+        return LatentsOutput.build(latents_name=name, latents=latents, seed=None)
+
+    @singledispatchmethod
+    @staticmethod
+    def _encode_to_tensor(vae: AutoencoderKL, image_tensor: torch.FloatTensor) -> torch.FloatTensor:
+        assert isinstance(vae, torch.nn.Module)
+        image_tensor_dist = vae.encode(image_tensor).latent_dist
+        latents: torch.Tensor = image_tensor_dist.sample().to(
+            dtype=vae.dtype
+        )  # FIXME: uses torch.randn. make reproducible!
+        return latents
+
+    @_encode_to_tensor.register
+    @staticmethod
+    def _(vae: AutoencoderTiny, image_tensor: torch.FloatTensor) -> torch.FloatTensor:
+        assert isinstance(vae, torch.nn.Module)
+        latents: torch.FloatTensor = vae.encode(image_tensor).latents
+        return latents

invokeai/app/invocations/infill.py

@@ -42,15 +42,16 @@ class InfillImageProcessorInvocation(BaseInvocation, WithMetadata, WithBoard):
         """Infill the image with the specified method"""
         pass
 
-    def load_image(self, context: InvocationContext) -> tuple[Image.Image, bool]:
+    def load_image(self) -> tuple[Image.Image, bool]:
         """Process the image to have an alpha channel before being infilled"""
-        image = context.images.get_pil(self.image.image_name)
+        image = self._context.images.get_pil(self.image.image_name)
         has_alpha = True if image.mode == "RGBA" else False
         return image, has_alpha
 
     def invoke(self, context: InvocationContext) -> ImageOutput:
+        self._context = context
         # Retrieve and process image to be infilled
-        input_image, has_alpha = self.load_image(context)
+        input_image, has_alpha = self.load_image()
 
         # If the input image has no alpha channel, return it
         if has_alpha is False:
@@ -133,8 +134,12 @@ class LaMaInfillInvocation(InfillImageProcessorInvocation):
     """Infills transparent areas of an image using the LaMa model"""
 
     def infill(self, image: Image.Image):
-        lama = LaMA()
-        return lama(image)
+        with self._context.models.load_remote_model(
+            source="https://github.com/Sanster/models/releases/download/add_big_lama/big-lama.pt",
+            loader=LaMA.load_jit_model,
+        ) as model:
+            lama = LaMA(model)
+            return lama(image)
 
 
 @invocation("infill_cv2", title="CV2 Infill", tags=["image", "inpaint"], category="inpaint", version="1.2.2")

invokeai/app/invocations/ip_adapter.py

@@ -5,7 +5,7 @@ from pydantic import BaseModel, Field, field_validator, model_validator
 from typing_extensions import Self
 
 from invokeai.app.invocations.baseinvocation import BaseInvocation, BaseInvocationOutput, invocation, invocation_output
-from invokeai.app.invocations.fields import FieldDescriptions, Input, InputField, OutputField, TensorField, UIType
+from invokeai.app.invocations.fields import FieldDescriptions, InputField, OutputField, TensorField, UIType
 from invokeai.app.invocations.model import ModelIdentifierField
 from invokeai.app.invocations.primitives import ImageField
 from invokeai.app.invocations.util import validate_begin_end_step, validate_weights
@@ -58,7 +58,7 @@ class IPAdapterOutput(BaseInvocationOutput):
 CLIP_VISION_MODEL_MAP = {"ViT-H": "ip_adapter_sd_image_encoder", "ViT-G": "ip_adapter_sdxl_image_encoder"}
 
 
-@invocation("ip_adapter", title="IP-Adapter", tags=["ip_adapter", "control"], category="ip_adapter", version="1.4.0")
+@invocation("ip_adapter", title="IP-Adapter", tags=["ip_adapter", "control"], category="ip_adapter", version="1.4.1")
 class IPAdapterInvocation(BaseInvocation):
     """Collects IP-Adapter info to pass to other nodes."""
 
@@ -67,7 +67,6 @@ class IPAdapterInvocation(BaseInvocation):
     ip_adapter_model: ModelIdentifierField = InputField(
         description="The IP-Adapter model.",
         title="IP-Adapter Model",
-        input=Input.Direct,
         ui_order=-1,
         ui_type=UIType.IPAdapterModel,
     )

invokeai/app/invocations/latents_to_image.py

@@ -0,0 +1,107 @@
+import torch
+from diffusers.image_processor import VaeImageProcessor
+from diffusers.models.attention_processor import (
+    AttnProcessor2_0,
+    LoRAAttnProcessor2_0,
+    LoRAXFormersAttnProcessor,
+    XFormersAttnProcessor,
+)
+from diffusers.models.autoencoders.autoencoder_kl import AutoencoderKL
+from diffusers.models.autoencoders.autoencoder_tiny import AutoencoderTiny
+from diffusers.models.unets.unet_2d_condition import UNet2DConditionModel
+
+from invokeai.app.invocations.baseinvocation import BaseInvocation, invocation
+from invokeai.app.invocations.constants import DEFAULT_PRECISION
+from invokeai.app.invocations.fields import (
+    FieldDescriptions,
+    Input,
+    InputField,
+    LatentsField,
+    WithBoard,
+    WithMetadata,
+)
+from invokeai.app.invocations.model import VAEField
+from invokeai.app.invocations.primitives import ImageOutput
+from invokeai.app.services.shared.invocation_context import InvocationContext
+from invokeai.backend.stable_diffusion import set_seamless
+from invokeai.backend.util.devices import TorchDevice
+
+
+@invocation(
+    "l2i",
+    title="Latents to Image",
+    tags=["latents", "image", "vae", "l2i"],
+    category="latents",
+    version="1.2.2",
+)
+class LatentsToImageInvocation(BaseInvocation, WithMetadata, WithBoard):
+    """Generates an image from latents."""
+
+    latents: LatentsField = InputField(
+        description=FieldDescriptions.latents,
+        input=Input.Connection,
+    )
+    vae: VAEField = InputField(
+        description=FieldDescriptions.vae,
+        input=Input.Connection,
+    )
+    tiled: bool = InputField(default=False, description=FieldDescriptions.tiled)
+    fp32: bool = InputField(default=DEFAULT_PRECISION == torch.float32, description=FieldDescriptions.fp32)
+
+    @torch.no_grad()
+    def invoke(self, context: InvocationContext) -> ImageOutput:
+        latents = context.tensors.load(self.latents.latents_name)
+
+        vae_info = context.models.load(self.vae.vae)
+        assert isinstance(vae_info.model, (UNet2DConditionModel, AutoencoderKL, AutoencoderTiny))
+        with set_seamless(vae_info.model, self.vae.seamless_axes), vae_info as vae:
+            assert isinstance(vae, torch.nn.Module)
+            latents = latents.to(vae.device)
+            if self.fp32:
+                vae.to(dtype=torch.float32)
+
+                use_torch_2_0_or_xformers = hasattr(vae.decoder, "mid_block") and isinstance(
+                    vae.decoder.mid_block.attentions[0].processor,
+                    (
+                        AttnProcessor2_0,
+                        XFormersAttnProcessor,
+                        LoRAXFormersAttnProcessor,
+                        LoRAAttnProcessor2_0,
+                    ),
+                )
+                # if xformers or torch_2_0 is used attention block does not need
+                # to be in float32 which can save lots of memory
+                if use_torch_2_0_or_xformers:
+                    vae.post_quant_conv.to(latents.dtype)
+                    vae.decoder.conv_in.to(latents.dtype)
+                    vae.decoder.mid_block.to(latents.dtype)
+                else:
+                    latents = latents.float()
+
+            else:
+                vae.to(dtype=torch.float16)
+                latents = latents.half()
+
+            if self.tiled or context.config.get().force_tiled_decode:
+                vae.enable_tiling()
+            else:
+                vae.disable_tiling()
+
+            # clear memory as vae decode can request a lot
+            TorchDevice.empty_cache()
+
+            with torch.inference_mode():
+                # copied from diffusers pipeline
+                latents = latents / vae.config.scaling_factor
+                image = vae.decode(latents, return_dict=False)[0]
+                image = (image / 2 + 0.5).clamp(0, 1)  # denormalize
+                # we always cast to float32 as this does not cause significant overhead and is compatible with bfloat16
+                np_image = image.cpu().permute(0, 2, 3, 1).float().numpy()
+
+                image = VaeImageProcessor.numpy_to_pil(np_image)[0]
+
+        TorchDevice.empty_cache()
+
+        image_dto = context.images.save(image=image)
+
+        return ImageOutput.build(image_dto)

invokeai/app/invocations/model.py

@@ -11,6 +11,7 @@ from invokeai.backend.model_manager.config import AnyModelConfig, BaseModelType,
 from .baseinvocation import (
     BaseInvocation,
     BaseInvocationOutput,
+    Classification,
     invocation,
     invocation_output,
 )
@@ -93,19 +94,46 @@ class ModelLoaderOutput(UNetOutput, CLIPOutput, VAEOutput):
     pass
 
 
+@invocation_output("model_identifier_output")
+class ModelIdentifierOutput(BaseInvocationOutput):
+    """Model identifier output"""
+
+    model: ModelIdentifierField = OutputField(description="Model identifier", title="Model")
+
+
+@invocation(
+    "model_identifier",
+    title="Model identifier",
+    tags=["model"],
+    category="model",
+    version="1.0.0",
+    classification=Classification.Prototype,
+)
+class ModelIdentifierInvocation(BaseInvocation):
+    """Selects any model, outputting it its identifier. Be careful with this one! The identifier will be accepted as
+    input for any model, even if the model types don't match. If you connect this to a mismatched input, you'll get an
+    error."""
+
+    model: ModelIdentifierField = InputField(description="The model to select", title="Model")
+
+    def invoke(self, context: InvocationContext) -> ModelIdentifierOutput:
+        if not context.models.exists(self.model.key):
+            raise Exception(f"Unknown model {self.model.key}")
+
+        return ModelIdentifierOutput(model=self.model)
+
+
 @invocation(
     "main_model_loader",
     title="Main Model",
     tags=["model"],
     category="model",
-    version="1.0.2",
+    version="1.0.3",
 )
 class MainModelLoaderInvocation(BaseInvocation):
     """Loads a main model, outputting its submodels."""
 
-    model: ModelIdentifierField = InputField(
-        description=FieldDescriptions.main_model, input=Input.Direct, ui_type=UIType.MainModel
-    )
+    model: ModelIdentifierField = InputField(description=FieldDescriptions.main_model, ui_type=UIType.MainModel)
     # TODO: precision?
 
     def invoke(self, context: InvocationContext) -> ModelLoaderOutput:
@@ -134,12 +162,12 @@ class LoRALoaderOutput(BaseInvocationOutput):
     clip: Optional[CLIPField] = OutputField(default=None, description=FieldDescriptions.clip, title="CLIP")
 
 
-@invocation("lora_loader", title="LoRA", tags=["model"], category="model", version="1.0.2")
+@invocation("lora_loader", title="LoRA", tags=["model"], category="model", version="1.0.3")
 class LoRALoaderInvocation(BaseInvocation):
     """Apply selected lora to unet and text_encoder."""
 
     lora: ModelIdentifierField = InputField(
-        description=FieldDescriptions.lora_model, input=Input.Direct, title="LoRA", ui_type=UIType.LoRAModel
+        description=FieldDescriptions.lora_model, title="LoRA", ui_type=UIType.LoRAModel
     )
     weight: float = InputField(default=0.75, description=FieldDescriptions.lora_weight)
     unet: Optional[UNetField] = InputField(
@@ -190,6 +218,75 @@ class LoRALoaderInvocation(BaseInvocation):
         return output
 
 
+@invocation_output("lora_selector_output")
+class LoRASelectorOutput(BaseInvocationOutput):
+    """Model loader output"""
+
+    lora: LoRAField = OutputField(description="LoRA model and weight", title="LoRA")
+
+
+@invocation("lora_selector", title="LoRA Selector", tags=["model"], category="model", version="1.0.1")
+class LoRASelectorInvocation(BaseInvocation):
+    """Selects a LoRA model and weight."""
+
+    lora: ModelIdentifierField = InputField(
+        description=FieldDescriptions.lora_model, title="LoRA", ui_type=UIType.LoRAModel
+    )
+    weight: float = InputField(default=0.75, description=FieldDescriptions.lora_weight)
+
+    def invoke(self, context: InvocationContext) -> LoRASelectorOutput:
+        return LoRASelectorOutput(lora=LoRAField(lora=self.lora, weight=self.weight))
+
+
+@invocation("lora_collection_loader", title="LoRA Collection Loader", tags=["model"], category="model", version="1.0.0")
+class LoRACollectionLoader(BaseInvocation):
+    """Applies a collection of LoRAs to the provided UNet and CLIP models."""
+
+    loras: LoRAField | list[LoRAField] = InputField(
+        description="LoRA models and weights. May be a single LoRA or collection.", title="LoRAs"
+    )
+    unet: Optional[UNetField] = InputField(
+        default=None,
+        description=FieldDescriptions.unet,
+        input=Input.Connection,
+        title="UNet",
+    )
+    clip: Optional[CLIPField] = InputField(
+        default=None,
+        description=FieldDescriptions.clip,
+        input=Input.Connection,
+        title="CLIP",
+    )
+
+    def invoke(self, context: InvocationContext) -> LoRALoaderOutput:
+        output = LoRALoaderOutput()
+        loras = self.loras if isinstance(self.loras, list) else [self.loras]
+        added_loras: list[str] = []
+
+        for lora in loras:
+            if lora.lora.key in added_loras:
+                continue
+
+            if not context.models.exists(lora.lora.key):
+                raise Exception(f"Unknown lora: {lora.lora.key}!")
+
+            assert lora.lora.base in (BaseModelType.StableDiffusion1, BaseModelType.StableDiffusion2)
+
+            added_loras.append(lora.lora.key)
+
+            if self.unet is not None:
+                if output.unet is None:
+                    output.unet = self.unet.model_copy(deep=True)
+                output.unet.loras.append(lora)
+
+            if self.clip is not None:
+                if output.clip is None:
+                    output.clip = self.clip.model_copy(deep=True)
+                output.clip.loras.append(lora)
+
+        return output
+
+
 @invocation_output("sdxl_lora_loader_output")
 class SDXLLoRALoaderOutput(BaseInvocationOutput):
     """SDXL LoRA Loader Output"""
@@ -204,13 +301,13 @@ class SDXLLoRALoaderOutput(BaseInvocationOutput):
     title="SDXL LoRA",
     tags=["lora", "model"],
     category="model",
-    version="1.0.2",
+    version="1.0.3",
 )
 class SDXLLoRALoaderInvocation(BaseInvocation):
     """Apply selected lora to unet and text_encoder."""
 
     lora: ModelIdentifierField = InputField(
-        description=FieldDescriptions.lora_model, input=Input.Direct, title="LoRA", ui_type=UIType.LoRAModel
+        description=FieldDescriptions.lora_model, title="LoRA", ui_type=UIType.LoRAModel
     )
     weight: float = InputField(default=0.75, description=FieldDescriptions.lora_weight)
     unet: Optional[UNetField] = InputField(
@@ -279,12 +376,78 @@ class SDXLLoRALoaderInvocation(BaseInvocation):
         return output
 
 
-@invocation("vae_loader", title="VAE", tags=["vae", "model"], category="model", version="1.0.2")
+@invocation(
+    "sdxl_lora_collection_loader",
+    title="SDXL LoRA Collection Loader",
+    tags=["model"],
+    category="model",
+    version="1.0.0",
+)
+class SDXLLoRACollectionLoader(BaseInvocation):
+    """Applies a collection of SDXL LoRAs to the provided UNet and CLIP models."""
+
+    loras: LoRAField | list[LoRAField] = InputField(
+        description="LoRA models and weights. May be a single LoRA or collection.", title="LoRAs"
+    )
+    unet: Optional[UNetField] = InputField(
+        default=None,
+        description=FieldDescriptions.unet,
+        input=Input.Connection,
+        title="UNet",
+    )
+    clip: Optional[CLIPField] = InputField(
+        default=None,
+        description=FieldDescriptions.clip,
+        input=Input.Connection,
+        title="CLIP",
+    )
+    clip2: Optional[CLIPField] = InputField(
+        default=None,
+        description=FieldDescriptions.clip,
+        input=Input.Connection,
+        title="CLIP 2",
+    )
+
+    def invoke(self, context: InvocationContext) -> SDXLLoRALoaderOutput:
+        output = SDXLLoRALoaderOutput()
+        loras = self.loras if isinstance(self.loras, list) else [self.loras]
+        added_loras: list[str] = []
+
+        for lora in loras:
+            if lora.lora.key in added_loras:
+                continue
+
+            if not context.models.exists(lora.lora.key):
+                raise Exception(f"Unknown lora: {lora.lora.key}!")
+
+            assert lora.lora.base is BaseModelType.StableDiffusionXL
+
+            added_loras.append(lora.lora.key)
+
+            if self.unet is not None:
+                if output.unet is None:
+                    output.unet = self.unet.model_copy(deep=True)
+                output.unet.loras.append(lora)
+
+            if self.clip is not None:
+                if output.clip is None:
+                    output.clip = self.clip.model_copy(deep=True)
+                output.clip.loras.append(lora)
+
+            if self.clip2 is not None:
+                if output.clip2 is None:
+                    output.clip2 = self.clip2.model_copy(deep=True)
+                output.clip2.loras.append(lora)
+
+        return output
+
+
+@invocation("vae_loader", title="VAE", tags=["vae", "model"], category="model", version="1.0.3")
 class VAELoaderInvocation(BaseInvocation):
     """Loads a VAE model, outputting a VaeLoaderOutput"""
 
     vae_model: ModelIdentifierField = InputField(
-        description=FieldDescriptions.vae_model, input=Input.Direct, title="VAE", ui_type=UIType.VAEModel
+        description=FieldDescriptions.vae_model, title="VAE", ui_type=UIType.VAEModel
     )
 
     def invoke(self, context: InvocationContext) -> VAEOutput:

invokeai/app/invocations/resize_latents.py

@@ -0,0 +1,103 @@
+from typing import Literal
+
+import torch
+
+from invokeai.app.invocations.baseinvocation import BaseInvocation, invocation
+from invokeai.app.invocations.constants import LATENT_SCALE_FACTOR
+from invokeai.app.invocations.fields import (
+    FieldDescriptions,
+    Input,
+    InputField,
+    LatentsField,
+)
+from invokeai.app.invocations.primitives import LatentsOutput
+from invokeai.app.services.shared.invocation_context import InvocationContext
+from invokeai.backend.util.devices import TorchDevice
+
+LATENTS_INTERPOLATION_MODE = Literal["nearest", "linear", "bilinear", "bicubic", "trilinear", "area", "nearest-exact"]
+
+
+@invocation(
+    "lresize",
+    title="Resize Latents",
+    tags=["latents", "resize"],
+    category="latents",
+    version="1.0.2",
+)
+class ResizeLatentsInvocation(BaseInvocation):
+    """Resizes latents to explicit width/height (in pixels). Provided dimensions are floor-divided by 8."""
+
+    latents: LatentsField = InputField(
+        description=FieldDescriptions.latents,
+        input=Input.Connection,
+    )
+    width: int = InputField(
+        ge=64,
+        multiple_of=LATENT_SCALE_FACTOR,
+        description=FieldDescriptions.width,
+    )
+    height: int = InputField(
+        ge=64,
+        multiple_of=LATENT_SCALE_FACTOR,
+        description=FieldDescriptions.width,
+    )
+    mode: LATENTS_INTERPOLATION_MODE = InputField(default="bilinear", description=FieldDescriptions.interp_mode)
+    antialias: bool = InputField(default=False, description=FieldDescriptions.torch_antialias)
+
+    def invoke(self, context: InvocationContext) -> LatentsOutput:
+        latents = context.tensors.load(self.latents.latents_name)
+        device = TorchDevice.choose_torch_device()
+
+        resized_latents = torch.nn.functional.interpolate(
+            latents.to(device),
+            size=(self.height // LATENT_SCALE_FACTOR, self.width // LATENT_SCALE_FACTOR),
+            mode=self.mode,
+            antialias=self.antialias if self.mode in ["bilinear", "bicubic"] else False,
+        )
+
+        # https://discuss.huggingface.co/t/memory-usage-by-later-pipeline-stages/23699
+        resized_latents = resized_latents.to("cpu")
+
+        TorchDevice.empty_cache()
+
+        name = context.tensors.save(tensor=resized_latents)
+        return LatentsOutput.build(latents_name=name, latents=resized_latents, seed=self.latents.seed)
+
+
+@invocation(
+    "lscale",
+    title="Scale Latents",
+    tags=["latents", "resize"],
+    category="latents",
+    version="1.0.2",
+)
+class ScaleLatentsInvocation(BaseInvocation):
+    """Scales latents by a given factor."""
+
+    latents: LatentsField = InputField(
+        description=FieldDescriptions.latents,
+        input=Input.Connection,
+    )
+    scale_factor: float = InputField(gt=0, description=FieldDescriptions.scale_factor)
+    mode: LATENTS_INTERPOLATION_MODE = InputField(default="bilinear", description=FieldDescriptions.interp_mode)
+    antialias: bool = InputField(default=False, description=FieldDescriptions.torch_antialias)
+
+    def invoke(self, context: InvocationContext) -> LatentsOutput:
+        latents = context.tensors.load(self.latents.latents_name)
+
+        device = TorchDevice.choose_torch_device()
+
+        # resizing
+        resized_latents = torch.nn.functional.interpolate(
+            latents.to(device),
+            scale_factor=self.scale_factor,
+            mode=self.mode,
+            antialias=self.antialias if self.mode in ["bilinear", "bicubic"] else False,
+        )
+
+        # https://discuss.huggingface.co/t/memory-usage-by-later-pipeline-stages/23699
+        resized_latents = resized_latents.to("cpu")
+        TorchDevice.empty_cache()
+
+        name = context.tensors.save(tensor=resized_latents)
+        return LatentsOutput.build(latents_name=name, latents=resized_latents, seed=self.latents.seed)

invokeai/app/invocations/scheduler.py

@@ -0,0 +1,34 @@
+from invokeai.app.invocations.baseinvocation import BaseInvocation, BaseInvocationOutput, invocation, invocation_output
+from invokeai.app.invocations.constants import SCHEDULER_NAME_VALUES
+from invokeai.app.invocations.fields import (
+    FieldDescriptions,
+    InputField,
+    OutputField,
+    UIType,
+)
+from invokeai.app.services.shared.invocation_context import InvocationContext
+
+
+@invocation_output("scheduler_output")
+class SchedulerOutput(BaseInvocationOutput):
+    scheduler: SCHEDULER_NAME_VALUES = OutputField(description=FieldDescriptions.scheduler, ui_type=UIType.Scheduler)
+
+
+@invocation(
+    "scheduler",
+    title="Scheduler",
+    tags=["scheduler"],
+    category="latents",
+    version="1.0.0",
+)
+class SchedulerInvocation(BaseInvocation):
+    """Selects a scheduler."""
+
+    scheduler: SCHEDULER_NAME_VALUES = InputField(
+        default="euler",
+        description=FieldDescriptions.scheduler,
+        ui_type=UIType.Scheduler,
+    )
+
+    def invoke(self, context: InvocationContext) -> SchedulerOutput:
+        return SchedulerOutput(scheduler=self.scheduler)

invokeai/app/invocations/sdxl.py

@@ -1,4 +1,4 @@
-from invokeai.app.invocations.fields import FieldDescriptions, Input, InputField, OutputField, UIType
+from invokeai.app.invocations.fields import FieldDescriptions, InputField, OutputField, UIType
 from invokeai.app.services.shared.invocation_context import InvocationContext
 from invokeai.backend.model_manager import SubModelType
 
@@ -30,12 +30,12 @@ class SDXLRefinerModelLoaderOutput(BaseInvocationOutput):
     vae: VAEField = OutputField(description=FieldDescriptions.vae, title="VAE")
 
 
-@invocation("sdxl_model_loader", title="SDXL Main Model", tags=["model", "sdxl"], category="model", version="1.0.2")
+@invocation("sdxl_model_loader", title="SDXL Main Model", tags=["model", "sdxl"], category="model", version="1.0.3")
 class SDXLModelLoaderInvocation(BaseInvocation):
     """Loads an sdxl base model, outputting its submodels."""
 
     model: ModelIdentifierField = InputField(
-        description=FieldDescriptions.sdxl_main_model, input=Input.Direct, ui_type=UIType.SDXLMainModel
+        description=FieldDescriptions.sdxl_main_model, ui_type=UIType.SDXLMainModel
     )
     # TODO: precision?
 
@@ -67,13 +67,13 @@ class SDXLModelLoaderInvocation(BaseInvocation):
     title="SDXL Refiner Model",
     tags=["model", "sdxl", "refiner"],
     category="model",
-    version="1.0.2",
+    version="1.0.3",
 )
 class SDXLRefinerModelLoaderInvocation(BaseInvocation):
     """Loads an sdxl refiner model, outputting its submodels."""
 
     model: ModelIdentifierField = InputField(
-        description=FieldDescriptions.sdxl_refiner_model, input=Input.Direct, ui_type=UIType.SDXLRefinerModel
+        description=FieldDescriptions.sdxl_refiner_model, ui_type=UIType.SDXLRefinerModel
     )
     # TODO: precision?
 

invokeai/app/invocations/t2i_adapter.py

@@ -8,7 +8,7 @@ from invokeai.app.invocations.baseinvocation import (
     invocation,
     invocation_output,
 )
-from invokeai.app.invocations.fields import FieldDescriptions, ImageField, Input, InputField, OutputField, UIType
+from invokeai.app.invocations.fields import FieldDescriptions, ImageField, InputField, OutputField, UIType
 from invokeai.app.invocations.model import ModelIdentifierField
 from invokeai.app.invocations.util import validate_begin_end_step, validate_weights
 from invokeai.app.services.shared.invocation_context import InvocationContext
@@ -45,7 +45,7 @@ class T2IAdapterOutput(BaseInvocationOutput):
 
 
 @invocation(
-    "t2i_adapter", title="T2I-Adapter", tags=["t2i_adapter", "control"], category="t2i_adapter", version="1.0.2"
+    "t2i_adapter", title="T2I-Adapter", tags=["t2i_adapter", "control"], category="t2i_adapter", version="1.0.3"
 )
 class T2IAdapterInvocation(BaseInvocation):
     """Collects T2I-Adapter info to pass to other nodes."""
@@ -55,7 +55,6 @@ class T2IAdapterInvocation(BaseInvocation):
     t2i_adapter_model: ModelIdentifierField = InputField(
         description="The T2I-Adapter model.",
         title="T2I-Adapter Model",
-        input=Input.Direct,
         ui_order=-1,
         ui_type=UIType.T2IAdapterModel,
     )

invokeai/app/invocations/upscale.py

@@ -1,5 +1,4 @@
 # Copyright (c) 2022 Kyle Schouviller (https://github.com/kyle0654) & the InvokeAI Team
-from pathlib import Path
 from typing import Literal
 
 import cv2
@@ -10,10 +9,8 @@ from pydantic import ConfigDict
 from invokeai.app.invocations.fields import ImageField
 from invokeai.app.invocations.primitives import ImageOutput
 from invokeai.app.services.shared.invocation_context import InvocationContext
-from invokeai.app.util.download_with_progress import download_with_progress_bar
 from invokeai.backend.image_util.basicsr.rrdbnet_arch import RRDBNet
 from invokeai.backend.image_util.realesrgan.realesrgan import RealESRGAN
-from invokeai.backend.util.devices import TorchDevice
 
 from .baseinvocation import BaseInvocation, invocation
 from .fields import InputField, WithBoard, WithMetadata
@@ -52,7 +49,6 @@ class ESRGANInvocation(BaseInvocation, WithMetadata, WithBoard):
 
         rrdbnet_model = None
         netscale = None
-        esrgan_model_path = None
 
         if self.model_name in [
             "RealESRGAN_x4plus.pth",
@@ -95,28 +91,25 @@ class ESRGANInvocation(BaseInvocation, WithMetadata, WithBoard):
             context.logger.error(msg)
             raise ValueError(msg)
 
-        esrgan_model_path = Path(context.config.get().models_path, f"core/upscaling/realesrgan/{self.model_name}")
-
-        # Downloads the ESRGAN model if it doesn't already exist
-        download_with_progress_bar(
-            name=self.model_name, url=ESRGAN_MODEL_URLS[self.model_name], dest_path=esrgan_model_path
+        loadnet = context.models.load_remote_model(
+            source=ESRGAN_MODEL_URLS[self.model_name],
         )
 
-        upscaler = RealESRGAN(
-            scale=netscale,
-            model_path=esrgan_model_path,
-            model=rrdbnet_model,
-            half=False,
-            tile=self.tile_size,
-        )
+        with loadnet as loadnet_model:
+            upscaler = RealESRGAN(
+                scale=netscale,
+                loadnet=loadnet_model,
+                model=rrdbnet_model,
+                half=False,
+                tile=self.tile_size,
+            )
 
-        # prepare image - Real-ESRGAN uses cv2 internally, and cv2 uses BGR vs RGB for PIL
-        # TODO: This strips the alpha... is that okay?
-        cv2_image = cv2.cvtColor(np.array(image.convert("RGB")), cv2.COLOR_RGB2BGR)
-        upscaled_image = upscaler.upscale(cv2_image)
-        pil_image = Image.fromarray(cv2.cvtColor(upscaled_image, cv2.COLOR_BGR2RGB)).convert("RGBA")
+            # prepare image - Real-ESRGAN uses cv2 internally, and cv2 uses BGR vs RGB for PIL
+            # TODO: This strips the alpha... is that okay?
+            cv2_image = cv2.cvtColor(np.array(image.convert("RGB")), cv2.COLOR_RGB2BGR)
+            upscaled_image = upscaler.upscale(cv2_image)
 
-        TorchDevice.empty_cache()
+            pil_image = Image.fromarray(cv2.cvtColor(upscaled_image, cv2.COLOR_BGR2RGB)).convert("RGBA")
 
         image_dto = context.images.save(image=pil_image)
 

invokeai/app/services/bulk_download/bulk_download_default.py

@@ -106,9 +106,7 @@ class BulkDownloadService(BulkDownloadBase):
         if self._invoker:
             assert bulk_download_id is not None
             self._invoker.services.events.emit_bulk_download_started(
-                bulk_download_id=bulk_download_id,
-                bulk_download_item_id=bulk_download_item_id,
-                bulk_download_item_name=bulk_download_item_name,
+                bulk_download_id, bulk_download_item_id, bulk_download_item_name
             )
 
     def _signal_job_completed(
@@ -118,10 +116,8 @@ class BulkDownloadService(BulkDownloadBase):
         if self._invoker:
             assert bulk_download_id is not None
             assert bulk_download_item_name is not None
-            self._invoker.services.events.emit_bulk_download_completed(
-                bulk_download_id=bulk_download_id,
-                bulk_download_item_id=bulk_download_item_id,
-                bulk_download_item_name=bulk_download_item_name,
+            self._invoker.services.events.emit_bulk_download_complete(
+                bulk_download_id, bulk_download_item_id, bulk_download_item_name
             )
 
     def _signal_job_failed(
@@ -131,11 +127,8 @@ class BulkDownloadService(BulkDownloadBase):
         if self._invoker:
             assert bulk_download_id is not None
             assert exception is not None
-            self._invoker.services.events.emit_bulk_download_failed(
-                bulk_download_id=bulk_download_id,
-                bulk_download_item_id=bulk_download_item_id,
-                bulk_download_item_name=bulk_download_item_name,
-                error=str(exception),
+            self._invoker.services.events.emit_bulk_download_error(
+                bulk_download_id, bulk_download_item_id, bulk_download_item_name, str(exception)
             )
 
     def stop(self, *args, **kwargs):

invokeai/app/services/config/config_default.py

@@ -26,13 +26,13 @@ LEGACY_INIT_FILE = Path("invokeai.init")
 DEFAULT_RAM_CACHE = 10.0
 DEFAULT_VRAM_CACHE = 0.25
 DEFAULT_CONVERT_CACHE = 20.0
-DEVICE = Literal["auto", "cpu", "cuda", "cuda:1", "mps"]
-PRECISION = Literal["auto", "float16", "bfloat16", "float32"]
+DEVICE = Literal["auto", "cpu", "cuda:0", "cuda:1", "cuda:2", "cuda:3", "cuda:4", "cuda:5", "cuda:6", "cuda:7", "mps"]
+PRECISION = Literal["auto", "float16", "bfloat16", "float32", "autocast"]
 ATTENTION_TYPE = Literal["auto", "normal", "xformers", "sliced", "torch-sdp"]
 ATTENTION_SLICE_SIZE = Literal["auto", "balanced", "max", 1, 2, 3, 4, 5, 6, 7, 8]
 LOG_FORMAT = Literal["plain", "color", "syslog", "legacy"]
 LOG_LEVEL = Literal["debug", "info", "warning", "error", "critical"]
-CONFIG_SCHEMA_VERSION = "4.0.1"
+CONFIG_SCHEMA_VERSION = "4.0.2"
 
 
 def get_default_ram_cache_size() -> float:
@@ -86,6 +86,7 @@ class InvokeAIAppConfig(BaseSettings):
         patchmatch: Enable patchmatch inpaint code.
         models_dir: Path to the models directory.
         convert_cache_dir: Path to the converted models cache directory. When loading a non-diffusers model, it will be converted and store on disk at this location.
+        download_cache_dir: Path to the directory that contains dynamically downloaded models.
         legacy_conf_dir: Path to directory of legacy checkpoint config files.
         db_dir: Path to InvokeAI databases directory.
         outputs_dir: Path to directory for outputs.
@@ -104,14 +105,17 @@ class InvokeAIAppConfig(BaseSettings):
         convert_cache: Maximum size of on-disk converted models cache (GB).
         lazy_offload: Keep models in VRAM until their space is needed.
         log_memory_usage: If True, a memory snapshot will be captured before and after every model cache operation, and the result will be logged (at debug level). There is a time cost to capturing the memory snapshots, so it is recommended to only enable this feature if you are actively inspecting the model cache's behaviour.
-        device: Preferred execution device. `auto` will choose the device depending on the hardware platform and the installed torch capabilities.<br>Valid values: `auto`, `cpu`, `cuda`, `cuda:1`, `mps`
-        precision: Floating point precision. `float16` will consume half the memory of `float32` but produce slightly lower-quality images. The `auto` setting will guess the proper precision based on your video card and operating system.<br>Valid values: `auto`, `float16`, `bfloat16`, `float32`
+        device: Preferred execution device. `auto` will choose the device depending on the hardware platform and the installed torch capabilities.<br>Valid values: `auto`, `cpu`, `cuda:0`, `cuda:1`, `cuda:2`, `cuda:3`, `cuda:4`, `cuda:5`, `cuda:6`, `cuda:7`, `mps`
+        devices: List of execution devices; will override default device selected.
+        precision: Floating point precision. `float16` will consume half the memory of `float32` but produce slightly lower-quality images. The `auto` setting will guess the proper precision based on your video card and operating system.<br>Valid values: `auto`, `float16`, `bfloat16`, `float32`, `autocast`
         sequential_guidance: Whether to calculate guidance in serial instead of in parallel, lowering memory requirements.
         attention_type: Attention type.<br>Valid values: `auto`, `normal`, `xformers`, `sliced`, `torch-sdp`
         attention_slice_size: Slice size, valid when attention_type=="sliced".<br>Valid values: `auto`, `balanced`, `max`, `1`, `2`, `3`, `4`, `5`, `6`, `7`, `8`
         force_tiled_decode: Whether to enable tiled VAE decode (reduces memory consumption with some performance penalty).
         pil_compress_level: The compress_level setting of PIL.Image.save(), used for PNG encoding. All settings are lossless. 0 = no compression, 1 = fastest with slightly larger filesize, 9 = slowest with smallest filesize. 1 is typically the best setting.
         max_queue_size: Maximum number of items in the session queue.
+        max_threads: Maximum number of session queue execution threads. Autocalculated from number of GPUs if not set.
+        clear_queue_on_startup: Empties session queue on startup.
         allow_nodes: List of nodes to allow. Omit to allow all.
         deny_nodes: List of nodes to deny. Omit to deny none.
         node_cache_size: How many cached nodes to keep in memory.
@@ -146,7 +150,8 @@ class InvokeAIAppConfig(BaseSettings):
 
     # PATHS
     models_dir:                    Path = Field(default=Path("models"),     description="Path to the models directory.")
-    convert_cache_dir:             Path = Field(default=Path("models/.cache"), description="Path to the converted models cache directory. When loading a non-diffusers model, it will be converted and store on disk at this location.")
+    convert_cache_dir:             Path = Field(default=Path("models/.convert_cache"), description="Path to the converted models cache directory. When loading a non-diffusers model, it will be converted and store on disk at this location.")
+    download_cache_dir:            Path = Field(default=Path("models/.download_cache"), description="Path to the directory that contains dynamically downloaded models.")
     legacy_conf_dir:               Path = Field(default=Path("configs"), description="Path to directory of legacy checkpoint config files.")
     db_dir:                        Path = Field(default=Path("databases"),  description="Path to InvokeAI databases directory.")
     outputs_dir:                   Path = Field(default=Path("outputs"),    description="Path to directory for outputs.")
@@ -175,6 +180,7 @@ class InvokeAIAppConfig(BaseSettings):
 
     # DEVICE
     device:                      DEVICE = Field(default="auto",             description="Preferred execution device. `auto` will choose the device depending on the hardware platform and the installed torch capabilities.")
+    devices:      Optional[list[DEVICE]] = Field(default=None,              description="List of execution devices; will override default device selected.")
     precision:                PRECISION = Field(default="auto",             description="Floating point precision. `float16` will consume half the memory of `float32` but produce slightly lower-quality images. The `auto` setting will guess the proper precision based on your video card and operating system.")
 
     # GENERATION
@@ -184,6 +190,8 @@ class InvokeAIAppConfig(BaseSettings):
     force_tiled_decode:            bool = Field(default=False,              description="Whether to enable tiled VAE decode (reduces memory consumption with some performance penalty).")
     pil_compress_level:             int = Field(default=1,                  description="The compress_level setting of PIL.Image.save(), used for PNG encoding. All settings are lossless. 0 = no compression, 1 = fastest with slightly larger filesize, 9 = slowest with smallest filesize. 1 is typically the best setting.")
     max_queue_size:                 int = Field(default=10000, gt=0,        description="Maximum number of items in the session queue.")
+    max_threads:          Optional[int] = Field(default=None,               description="Maximum number of session queue execution threads. Autocalculated from number of GPUs if not set.")
+    clear_queue_on_startup:        bool = Field(default=False,              description="Empties session queue on startup.")
 
     # NODES
     allow_nodes:    Optional[list[str]] = Field(default=None,               description="List of nodes to allow. Omit to allow all.")
@@ -303,6 +311,11 @@ class InvokeAIAppConfig(BaseSettings):
         """Path to the converted cache models directory, resolved to an absolute path.."""
         return self._resolve(self.convert_cache_dir)
 
+    @property
+    def download_cache_path(self) -> Path:
+        """Path to the downloaded models directory, resolved to an absolute path.."""
+        return self._resolve(self.download_cache_dir)
+
     @property
     def custom_nodes_path(self) -> Path:
         """Path to the custom nodes directory, resolved to an absolute path.."""
@@ -367,9 +380,6 @@ def migrate_v3_config_dict(config_dict: dict[str, Any]) -> InvokeAIAppConfig:
             # `max_cache_size` was renamed to `ram` some time in v3, but both names were used
             if k == "max_cache_size" and "ram" not in category_dict:
                 parsed_config_dict["ram"] = v
-            # `max_vram_cache_size` was renamed to `vram` some time in v3, but both names were used
-            if k == "max_vram_cache_size" and "vram" not in category_dict:
-                parsed_config_dict["vram"] = v
             # autocast was removed in v4.0.1
             if k == "precision" and v == "autocast":
                 parsed_config_dict["precision"] = "auto"
@@ -417,6 +427,27 @@ def migrate_v4_0_0_config_dict(config_dict: dict[str, Any]) -> InvokeAIAppConfig
     return config
 
 
+def migrate_v4_0_1_config_dict(config_dict: dict[str, Any]) -> InvokeAIAppConfig:
+    """Migrate v4.0.1 config dictionary to a current config object.
+
+    A few new multi-GPU options were added in 4.0.2, and this simply
+    updates the schema label.
+
+    Args:
+        config_dict: A dictionary of settings from a v4.0.1 config file.
+
+    Returns:
+        An instance of `InvokeAIAppConfig` with the migrated settings.
+    """
+    parsed_config_dict: dict[str, Any] = {}
+    for k, _ in config_dict.items():
+        if k == "schema_version":
+            parsed_config_dict[k] = CONFIG_SCHEMA_VERSION
+    config = DefaultInvokeAIAppConfig.model_validate(parsed_config_dict)
+    return config
+
+
+# TO DO: replace this with a formal registration and migration system
 def load_and_migrate_config(config_path: Path) -> InvokeAIAppConfig:
     """Load and migrate a config file to the latest version.
 
@@ -448,6 +479,10 @@ def load_and_migrate_config(config_path: Path) -> InvokeAIAppConfig:
         loaded_config_dict = migrate_v4_0_0_config_dict(loaded_config_dict)
         loaded_config_dict.write_file(config_path)
 
+    elif loaded_config_dict["schema_version"] == "4.0.1":
+        loaded_config_dict = migrate_v4_0_1_config_dict(loaded_config_dict)
+        loaded_config_dict.write_file(config_path)
+
     # Attempt to load as a v4 config file
     try:
         # Meta is not included in the model fields, so we need to validate it separately

invokeai/app/services/download/__init__.py

@@ -1,10 +1,17 @@
 """Init file for download queue."""
 
-from .download_base import DownloadJob, DownloadJobStatus, DownloadQueueServiceBase, UnknownJobIDException
+from .download_base import (
+    DownloadJob,
+    DownloadJobStatus,
+    DownloadQueueServiceBase,
+    MultiFileDownloadJob,
+    UnknownJobIDException,
+)
 from .download_default import DownloadQueueService, TqdmProgress
 
 __all__ = [
     "DownloadJob",
+    "MultiFileDownloadJob",
     "DownloadQueueServiceBase",
     "DownloadQueueService",
     "TqdmProgress",

invokeai/app/services/download/download_base.py

@@ -5,11 +5,13 @@ from abc import ABC, abstractmethod
 from enum import Enum
 from functools import total_ordering
 from pathlib import Path
-from typing import Any, Callable, List, Optional
+from typing import Any, Callable, List, Optional, Set, Union
 
 from pydantic import BaseModel, Field, PrivateAttr
 from pydantic.networks import AnyHttpUrl
 
+from invokeai.backend.model_manager.metadata import RemoteModelFile
+
 
 class DownloadJobStatus(str, Enum):
     """State of a download job."""
@@ -33,30 +35,23 @@ class ServiceInactiveException(Exception):
     """This exception is raised when user attempts to initiate a download before the service is started."""
 
 
-DownloadEventHandler = Callable[["DownloadJob"], None]
-DownloadExceptionHandler = Callable[["DownloadJob", Optional[Exception]], None]
+SingleFileDownloadEventHandler = Callable[["DownloadJob"], None]
+SingleFileDownloadExceptionHandler = Callable[["DownloadJob", Optional[Exception]], None]
+MultiFileDownloadEventHandler = Callable[["MultiFileDownloadJob"], None]
+MultiFileDownloadExceptionHandler = Callable[["MultiFileDownloadJob", Optional[Exception]], None]
+DownloadEventHandler = Union[SingleFileDownloadEventHandler, MultiFileDownloadEventHandler]
+DownloadExceptionHandler = Union[SingleFileDownloadExceptionHandler, MultiFileDownloadExceptionHandler]
 
 
-@total_ordering
-class DownloadJob(BaseModel):
-    """Class to monitor and control a model download request."""
+class DownloadJobBase(BaseModel):
+    """Base of classes to monitor and control downloads."""
 
-    # required variables to be passed in on creation
-    source: AnyHttpUrl = Field(description="Where to download from. Specific types specified in child classes.")
-    dest: Path = Field(description="Destination of downloaded model on local disk; a directory or file path")
-    access_token: Optional[str] = Field(default=None, description="authorization token for protected resources")
     # automatically assigned on creation
     id: int = Field(description="Numeric ID of this job", default=-1)  # default id is a sentinel
-    priority: int = Field(default=10, description="Queue priority; lower values are higher priority")
 
-    # set internally during download process
+    dest: Path = Field(description="Initial destination of downloaded model on local disk; a directory or file path")
+    download_path: Optional[Path] = Field(default=None, description="Final location of downloaded file or directory")
     status: DownloadJobStatus = Field(default=DownloadJobStatus.WAITING, description="Status of the download")
-    download_path: Optional[Path] = Field(default=None, description="Final location of downloaded file")
-    job_started: Optional[str] = Field(default=None, description="Timestamp for when the download job started")
-    job_ended: Optional[str] = Field(
-        default=None, description="Timestamp for when the download job ende1d (completed or errored)"
-    )
-    content_type: Optional[str] = Field(default=None, description="Content type of downloaded file")
     bytes: int = Field(default=0, description="Bytes downloaded so far")
     total_bytes: int = Field(default=0, description="Total file size (bytes)")
 
@@ -74,14 +69,6 @@ class DownloadJob(BaseModel):
     _on_cancelled: Optional[DownloadEventHandler] = PrivateAttr(default=None)
     _on_error: Optional[DownloadExceptionHandler] = PrivateAttr(default=None)
 
-    def __hash__(self) -> int:
-        """Return hash of the string representation of this object, for indexing."""
-        return hash(str(self))
-
-    def __le__(self, other: "DownloadJob") -> bool:
-        """Return True if this job's priority is less than another's."""
-        return self.priority <= other.priority
-
     def cancel(self) -> None:
         """Call to cancel the job."""
         self._cancelled = True
@@ -98,6 +85,11 @@ class DownloadJob(BaseModel):
         """Return true if job completed without errors."""
         return self.status == DownloadJobStatus.COMPLETED
 
+    @property
+    def waiting(self) -> bool:
+        """Return true if the job is waiting to run."""
+        return self.status == DownloadJobStatus.WAITING
+
     @property
     def running(self) -> bool:
         """Return true if the job is running."""
@@ -154,6 +146,37 @@ class DownloadJob(BaseModel):
         self._on_cancelled = on_cancelled
 
 
+@total_ordering
+class DownloadJob(DownloadJobBase):
+    """Class to monitor and control a model download request."""
+
+    # required variables to be passed in on creation
+    source: AnyHttpUrl = Field(description="Where to download from. Specific types specified in child classes.")
+    access_token: Optional[str] = Field(default=None, description="authorization token for protected resources")
+    priority: int = Field(default=10, description="Queue priority; lower values are higher priority")
+
+    # set internally during download process
+    job_started: Optional[str] = Field(default=None, description="Timestamp for when the download job started")
+    job_ended: Optional[str] = Field(
+        default=None, description="Timestamp for when the download job ende1d (completed or errored)"
+    )
+    content_type: Optional[str] = Field(default=None, description="Content type of downloaded file")
+
+    def __hash__(self) -> int:
+        """Return hash of the string representation of this object, for indexing."""
+        return hash(str(self))
+
+    def __le__(self, other: "DownloadJob") -> bool:
+        """Return True if this job's priority is less than another's."""
+        return self.priority <= other.priority
+
+
+class MultiFileDownloadJob(DownloadJobBase):
+    """Class to monitor and control multifile downloads."""
+
+    download_parts: Set[DownloadJob] = Field(default_factory=set, description="List of download parts.")
+
+
 class DownloadQueueServiceBase(ABC):
     """Multithreaded queue for downloading models via URL."""
 
@@ -201,6 +224,48 @@ class DownloadQueueServiceBase(ABC):
         """
         pass
 
+    @abstractmethod
+    def multifile_download(
+        self,
+        parts: List[RemoteModelFile],
+        dest: Path,
+        access_token: Optional[str] = None,
+        submit_job: bool = True,
+        on_start: Optional[DownloadEventHandler] = None,
+        on_progress: Optional[DownloadEventHandler] = None,
+        on_complete: Optional[DownloadEventHandler] = None,
+        on_cancelled: Optional[DownloadEventHandler] = None,
+        on_error: Optional[DownloadExceptionHandler] = None,
+    ) -> MultiFileDownloadJob:
+        """
+        Create and enqueue a multifile download job.
+
+        :param parts: Set of URL / filename pairs
+        :param dest: Path to download to. See below.
+        :param access_token: Access token to download the indicated files. If not provided,
+         each file's URL may be matched to an access token using the config file matching
+         system.
+        :param submit_job: If true [default] then submit the job for execution. Otherwise,
+         you will need to pass the job to submit_multifile_download().
+        :param on_start, on_progress, on_complete, on_error: Callbacks for the indicated
+         events.
+        :returns: A MultiFileDownloadJob object for monitoring the state of the download.
+
+        The `dest` argument is a Path object pointing to a directory. All downloads
+        with be placed inside this directory. The callbacks will receive the
+        MultiFileDownloadJob.
+        """
+        pass
+
+    @abstractmethod
+    def submit_multifile_download(self, job: MultiFileDownloadJob) -> None:
+        """
+        Enqueue a previously-created multi-file download job.
+
+        :param job: A MultiFileDownloadJob created with multifile_download()
+        """
+        pass
+
     @abstractmethod
     def submit_download_job(
         self,
@@ -252,7 +317,7 @@ class DownloadQueueServiceBase(ABC):
         pass
 
     @abstractmethod
-    def cancel_job(self, job: DownloadJob) -> None:
+    def cancel_job(self, job: DownloadJobBase) -> None:
         """Cancel the job, clearing partial downloads and putting it into ERROR state."""
         pass
 
@@ -262,7 +327,7 @@ class DownloadQueueServiceBase(ABC):
         pass
 
     @abstractmethod
-    def wait_for_job(self, job: DownloadJob, timeout: int = 0) -> DownloadJob:
+    def wait_for_job(self, job: DownloadJobBase, timeout: int = 0) -> DownloadJobBase:
         """Wait until the indicated download job has reached a terminal state.
 
         This will block until the indicated install job has completed,

invokeai/app/services/download/download_default.py

@@ -8,24 +8,28 @@ import time
 import traceback
 from pathlib import Path
 from queue import Empty, PriorityQueue
-from typing import Any, Dict, List, Optional, Set
+from typing import Any, Dict, List, Literal, Optional, Set
 
 import requests
 from pydantic.networks import AnyHttpUrl
 from requests import HTTPError
 from tqdm import tqdm
 
+from invokeai.app.services.config import InvokeAIAppConfig, get_config
 from invokeai.app.services.events.events_base import EventServiceBase
 from invokeai.app.util.misc import get_iso_timestamp
+from invokeai.backend.model_manager.metadata import RemoteModelFile
 from invokeai.backend.util.logging import InvokeAILogger
 
 from .download_base import (
     DownloadEventHandler,
     DownloadExceptionHandler,
     DownloadJob,
+    DownloadJobBase,
     DownloadJobCancelledException,
     DownloadJobStatus,
     DownloadQueueServiceBase,
+    MultiFileDownloadJob,
     ServiceInactiveException,
     UnknownJobIDException,
 )
@@ -40,20 +44,24 @@ class DownloadQueueService(DownloadQueueServiceBase):
     def __init__(
         self,
         max_parallel_dl: int = 5,
-        event_bus: Optional[EventServiceBase] = None,
+        app_config: Optional[InvokeAIAppConfig] = None,
+        event_bus: Optional["EventServiceBase"] = None,
         requests_session: Optional[requests.sessions.Session] = None,
     ):
         """
         Initialize DownloadQueue.
 
+        :param app_config: InvokeAIAppConfig object
         :param max_parallel_dl: Number of simultaneous downloads allowed [5].
         :param requests_session: Optional requests.sessions.Session object, for unit tests.
         """
+        self._app_config = app_config or get_config()
         self._jobs: Dict[int, DownloadJob] = {}
+        self._download_part2parent: Dict[AnyHttpUrl, MultiFileDownloadJob] = {}
         self._next_job_id = 0
         self._queue: PriorityQueue[DownloadJob] = PriorityQueue()
         self._stop_event = threading.Event()
-        self._job_completed_event = threading.Event()
+        self._job_terminated_event = threading.Event()
         self._worker_pool: Set[threading.Thread] = set()
         self._lock = threading.Lock()
         self._logger = InvokeAILogger.get_logger("DownloadQueueService")
@@ -105,18 +113,16 @@ class DownloadQueueService(DownloadQueueServiceBase):
             raise ServiceInactiveException(
                 "The download service is not currently accepting requests. Please call start() to initialize the service."
             )
-        with self._lock:
-            job.id = self._next_job_id
-            self._next_job_id += 1
-            job.set_callbacks(
-                on_start=on_start,
-                on_progress=on_progress,
-                on_complete=on_complete,
-                on_cancelled=on_cancelled,
-                on_error=on_error,
-            )
-            self._jobs[job.id] = job
-            self._queue.put(job)
+        job.id = self._next_id()
+        job.set_callbacks(
+            on_start=on_start,
+            on_progress=on_progress,
+            on_complete=on_complete,
+            on_cancelled=on_cancelled,
+            on_error=on_error,
+        )
+        self._jobs[job.id] = job
+        self._queue.put(job)
 
     def download(
         self,
@@ -139,7 +145,7 @@ class DownloadQueueService(DownloadQueueServiceBase):
             source=source,
             dest=dest,
             priority=priority,
-            access_token=access_token,
+            access_token=access_token or self._lookup_access_token(source),
         )
         self.submit_download_job(
             job,
@@ -151,10 +157,63 @@ class DownloadQueueService(DownloadQueueServiceBase):
         )
         return job
 
+    def multifile_download(
+        self,
+        parts: List[RemoteModelFile],
+        dest: Path,
+        access_token: Optional[str] = None,
+        submit_job: bool = True,
+        on_start: Optional[DownloadEventHandler] = None,
+        on_progress: Optional[DownloadEventHandler] = None,
+        on_complete: Optional[DownloadEventHandler] = None,
+        on_cancelled: Optional[DownloadEventHandler] = None,
+        on_error: Optional[DownloadExceptionHandler] = None,
+    ) -> MultiFileDownloadJob:
+        mfdj = MultiFileDownloadJob(dest=dest, id=self._next_id())
+        mfdj.set_callbacks(
+            on_start=on_start,
+            on_progress=on_progress,
+            on_complete=on_complete,
+            on_cancelled=on_cancelled,
+            on_error=on_error,
+        )
+
+        for part in parts:
+            url = part.url
+            path = dest / part.path
+            assert path.is_relative_to(dest), "only relative download paths accepted"
+            job = DownloadJob(
+                source=url,
+                dest=path,
+                access_token=access_token,
+            )
+            mfdj.download_parts.add(job)
+            self._download_part2parent[job.source] = mfdj
+        if submit_job:
+            self.submit_multifile_download(mfdj)
+        return mfdj
+
+    def submit_multifile_download(self, job: MultiFileDownloadJob) -> None:
+        for download_job in job.download_parts:
+            self.submit_download_job(
+                download_job,
+                on_start=self._mfd_started,
+                on_progress=self._mfd_progress,
+                on_complete=self._mfd_complete,
+                on_cancelled=self._mfd_cancelled,
+                on_error=self._mfd_error,
+            )
+
     def join(self) -> None:
         """Wait for all jobs to complete."""
         self._queue.join()
 
+    def _next_id(self) -> int:
+        with self._lock:
+            id = self._next_job_id
+            self._next_job_id += 1
+        return id
+
     def list_jobs(self) -> List[DownloadJob]:
         """List all the jobs."""
         return list(self._jobs.values())
@@ -176,14 +235,14 @@ class DownloadQueueService(DownloadQueueServiceBase):
         except KeyError as excp:
             raise UnknownJobIDException("Unrecognized job") from excp
 
-    def cancel_job(self, job: DownloadJob) -> None:
+    def cancel_job(self, job: DownloadJobBase) -> None:
         """
         Cancel the indicated job.
 
         If it is running it will be stopped.
         job.status will be set to DownloadJobStatus.CANCELLED
         """
-        with self._lock:
+        if job.status in [DownloadJobStatus.WAITING, DownloadJobStatus.RUNNING]:
             job.cancel()
 
     def cancel_all_jobs(self) -> None:
@@ -192,12 +251,12 @@ class DownloadQueueService(DownloadQueueServiceBase):
             if not job.in_terminal_state:
                 self.cancel_job(job)
 
-    def wait_for_job(self, job: DownloadJob, timeout: int = 0) -> DownloadJob:
+    def wait_for_job(self, job: DownloadJobBase, timeout: int = 0) -> DownloadJobBase:
         """Block until the indicated job has reached terminal state, or when timeout limit reached."""
         start = time.time()
         while not job.in_terminal_state:
-            if self._job_completed_event.wait(timeout=0.25):  # in case we miss an event
-                self._job_completed_event.clear()
+            if self._job_terminated_event.wait(timeout=0.25):  # in case we miss an event
+                self._job_terminated_event.clear()
             if timeout > 0 and time.time() - start > timeout:
                 raise TimeoutError("Timeout exceeded")
         return job
@@ -226,22 +285,25 @@ class DownloadQueueService(DownloadQueueServiceBase):
                 job.job_started = get_iso_timestamp()
                 self._do_download(job)
                 self._signal_job_complete(job)
-            except (OSError, HTTPError) as excp:
-                job.error_type = excp.__class__.__name__ + f"({str(excp)})"
-                job.error = traceback.format_exc()
-                self._signal_job_error(job, excp)
             except DownloadJobCancelledException:
                 self._signal_job_cancelled(job)
                 self._cleanup_cancelled_job(job)
-
+            except Exception as excp:
+                job.error_type = excp.__class__.__name__ + f"({str(excp)})"
+                job.error = traceback.format_exc()
+                self._signal_job_error(job, excp)
             finally:
                 job.job_ended = get_iso_timestamp()
-                self._job_completed_event.set()  # signal a change to terminal state
+                self._job_terminated_event.set()  # signal a change to terminal state
+                self._download_part2parent.pop(job.source, None)  # if this is a subpart of a multipart job, remove it
+                self._job_terminated_event.set()
                 self._queue.task_done()
+
         self._logger.debug(f"Download queue worker thread {threading.current_thread().name} exiting.")
 
     def _do_download(self, job: DownloadJob) -> None:
         """Do the actual download."""
+
         url = job.source
         header = {"Authorization": f"Bearer {job.access_token}"} if job.access_token else {}
         open_mode = "wb"
@@ -333,79 +395,53 @@ class DownloadQueueService(DownloadQueueServiceBase):
     def _in_progress_path(self, path: Path) -> Path:
         return path.with_name(path.name + ".downloading")
 
+    def _lookup_access_token(self, source: AnyHttpUrl) -> Optional[str]:
+        # Pull the token from config if it exists and matches the URL
+        token = None
+        for pair in self._app_config.remote_api_tokens or []:
+            if re.search(pair.url_regex, str(source)):
+                token = pair.token
+                break
+        return token
+
     def _signal_job_started(self, job: DownloadJob) -> None:
         job.status = DownloadJobStatus.RUNNING
-        if job.on_start:
-            try:
-                job.on_start(job)
-            except Exception as e:
-                self._logger.error(
-                    f"An error occurred while processing the on_start callback: {traceback.format_exception(e)}"
-                )
+        self._execute_cb(job, "on_start")
         if self._event_bus:
-            assert job.download_path
-            self._event_bus.emit_download_started(str(job.source), job.download_path.as_posix())
+            self._event_bus.emit_download_started(job)
 
     def _signal_job_progress(self, job: DownloadJob) -> None:
-        if job.on_progress:
-            try:
-                job.on_progress(job)
-            except Exception as e:
-                self._logger.error(
-                    f"An error occurred while processing the on_progress callback: {traceback.format_exception(e)}"
-                )
+        self._execute_cb(job, "on_progress")
         if self._event_bus:
-            assert job.download_path
-            self._event_bus.emit_download_progress(
-                str(job.source),
-                download_path=job.download_path.as_posix(),
-                current_bytes=job.bytes,
-                total_bytes=job.total_bytes,
-            )
+            self._event_bus.emit_download_progress(job)
 
     def _signal_job_complete(self, job: DownloadJob) -> None:
         job.status = DownloadJobStatus.COMPLETED
-        if job.on_complete:
-            try:
-                job.on_complete(job)
-            except Exception as e:
-                self._logger.error(
-                    f"An error occurred while processing the on_complete callback: {traceback.format_exception(e)}"
-                )
+        self._execute_cb(job, "on_complete")
         if self._event_bus:
-            assert job.download_path
-            self._event_bus.emit_download_complete(
-                str(job.source), download_path=job.download_path.as_posix(), total_bytes=job.total_bytes
-            )
+            self._event_bus.emit_download_complete(job)
 
     def _signal_job_cancelled(self, job: DownloadJob) -> None:
         if job.status not in [DownloadJobStatus.RUNNING, DownloadJobStatus.WAITING]:
             return
         job.status = DownloadJobStatus.CANCELLED
-        if job.on_cancelled:
-            try:
-                job.on_cancelled(job)
-            except Exception as e:
-                self._logger.error(
-                    f"An error occurred while processing the on_cancelled callback: {traceback.format_exception(e)}"
-                )
+        self._execute_cb(job, "on_cancelled")
         if self._event_bus:
-            self._event_bus.emit_download_cancelled(str(job.source))
+            self._event_bus.emit_download_cancelled(job)
+
+        # if multifile download, then signal the parent
+        if parent_job := self._download_part2parent.get(job.source, None):
+            if not parent_job.in_terminal_state:
+                parent_job.status = DownloadJobStatus.CANCELLED
+                self._execute_cb(parent_job, "on_cancelled")
 
     def _signal_job_error(self, job: DownloadJob, excp: Optional[Exception] = None) -> None:
         job.status = DownloadJobStatus.ERROR
         self._logger.error(f"{str(job.source)}: {traceback.format_exception(excp)}")
-        if job.on_error:
-            try:
-                job.on_error(job, excp)
-            except Exception as e:
-                self._logger.error(
-                    f"An error occurred while processing the on_error callback: {traceback.format_exception(e)}"
-                )
+        self._execute_cb(job, "on_error", excp)
+
         if self._event_bus:
-            assert job.error_type
-            assert job.error
-            self._event_bus.emit_download_error(str(job.source), error_type=job.error_type, error=job.error)
+            self._event_bus.emit_download_error(job)
 
     def _cleanup_cancelled_job(self, job: DownloadJob) -> None:
         self._logger.debug(f"Cleaning up leftover files from cancelled download job {job.download_path}")
@@ -416,6 +452,97 @@ class DownloadQueueService(DownloadQueueServiceBase):
         except OSError as excp:
             self._logger.warning(excp)
 
+    ########################################
+    # callbacks used for multifile downloads
+    ########################################
+    def _mfd_started(self, download_job: DownloadJob) -> None:
+        self._logger.info(f"File download started: {download_job.source}")
+        with self._lock:
+            mf_job = self._download_part2parent[download_job.source]
+            if mf_job.waiting:
+                mf_job.total_bytes = sum(x.total_bytes for x in mf_job.download_parts)
+                mf_job.status = DownloadJobStatus.RUNNING
+                assert download_job.download_path is not None
+                path_relative_to_destdir = download_job.download_path.relative_to(mf_job.dest)
+                mf_job.download_path = (
+                    mf_job.dest / path_relative_to_destdir.parts[0]
+                )  # keep just the first component of the path
+                self._execute_cb(mf_job, "on_start")
+
+    def _mfd_progress(self, download_job: DownloadJob) -> None:
+        with self._lock:
+            mf_job = self._download_part2parent[download_job.source]
+            if mf_job.cancelled:
+                for part in mf_job.download_parts:
+                    self.cancel_job(part)
+            elif mf_job.running:
+                mf_job.total_bytes = sum(x.total_bytes for x in mf_job.download_parts)
+                mf_job.bytes = sum(x.total_bytes for x in mf_job.download_parts)
+                self._execute_cb(mf_job, "on_progress")
+
+    def _mfd_complete(self, download_job: DownloadJob) -> None:
+        self._logger.info(f"Download complete: {download_job.source}")
+        with self._lock:
+            mf_job = self._download_part2parent[download_job.source]
+
+            # are there any more active jobs left in this task?
+            if mf_job.running and all(x.complete for x in mf_job.download_parts):
+                mf_job.status = DownloadJobStatus.COMPLETED
+                self._execute_cb(mf_job, "on_complete")
+
+            # we're done with this sub-job
+            self._job_terminated_event.set()
+
+    def _mfd_cancelled(self, download_job: DownloadJob) -> None:
+        with self._lock:
+            mf_job = self._download_part2parent[download_job.source]
+            assert mf_job is not None
+
+            if not mf_job.in_terminal_state:
+                self._logger.warning(f"Download cancelled: {download_job.source}")
+                mf_job.cancel()
+
+            for s in mf_job.download_parts:
+                self.cancel_job(s)
+
+    def _mfd_error(self, download_job: DownloadJob, excp: Optional[Exception] = None) -> None:
+        with self._lock:
+            mf_job = self._download_part2parent[download_job.source]
+            assert mf_job is not None
+            if not mf_job.in_terminal_state:
+                mf_job.status = download_job.status
+                mf_job.error = download_job.error
+                mf_job.error_type = download_job.error_type
+                self._execute_cb(mf_job, "on_error", excp)
+                self._logger.error(
+                    f"Cancelling {mf_job.dest} due to an error while downloading {download_job.source}: {str(excp)}"
+                )
+                for s in [x for x in mf_job.download_parts if x.running]:
+                    self.cancel_job(s)
+                self._download_part2parent.pop(download_job.source)
+                self._job_terminated_event.set()
+
+    def _execute_cb(
+        self,
+        job: DownloadJob | MultiFileDownloadJob,
+        callback_name: Literal[
+            "on_start",
+            "on_progress",
+            "on_complete",
+            "on_cancelled",
+            "on_error",
+        ],
+        excp: Optional[Exception] = None,
+    ) -> None:
+        if callback := getattr(job, callback_name, None):
+            args = [job, excp] if excp else [job]
+            try:
+                callback(*args)
+            except Exception as e:
+                self._logger.error(
+                    f"An error occurred while processing the {callback_name} callback: {traceback.format_exception(e)}"
+                )
+
 
 def get_pc_name_max(directory: str) -> int:
     if hasattr(os, "pathconf"):

invokeai/app/services/events/events_base.py

@@ -1,486 +1,199 @@
 # Copyright (c) 2022 Kyle Schouviller (https://github.com/kyle0654)
 
 
-from typing import Any, Dict, List, Optional, Union
+from typing import TYPE_CHECKING, Optional
 
-from invokeai.app.services.session_processor.session_processor_common import ProgressImage
-from invokeai.app.services.session_queue.session_queue_common import (
-    BatchStatus,
-    EnqueueBatchResult,
-    SessionQueueItem,
-    SessionQueueStatus,
+from invokeai.app.services.events.events_common import (
+    BatchEnqueuedEvent,
+    BulkDownloadCompleteEvent,
+    BulkDownloadErrorEvent,
+    BulkDownloadStartedEvent,
+    DownloadCancelledEvent,
+    DownloadCompleteEvent,
+    DownloadErrorEvent,
+    DownloadProgressEvent,
+    DownloadStartedEvent,
+    EventBase,
+    InvocationCompleteEvent,
+    InvocationDenoiseProgressEvent,
+    InvocationErrorEvent,
+    InvocationStartedEvent,
+    ModelInstallCancelledEvent,
+    ModelInstallCompleteEvent,
+    ModelInstallDownloadProgressEvent,
+    ModelInstallDownloadsCompleteEvent,
+    ModelInstallDownloadStartedEvent,
+    ModelInstallErrorEvent,
+    ModelInstallStartedEvent,
+    ModelLoadCompleteEvent,
+    ModelLoadStartedEvent,
+    QueueClearedEvent,
+    QueueItemStatusChangedEvent,
 )
-from invokeai.app.util.misc import get_timestamp
-from invokeai.backend.model_manager import AnyModelConfig
-from invokeai.backend.model_manager.config import SubModelType
+from invokeai.backend.stable_diffusion.diffusers_pipeline import PipelineIntermediateState
+
+if TYPE_CHECKING:
+    from invokeai.app.invocations.baseinvocation import BaseInvocation, BaseInvocationOutput
+    from invokeai.app.services.download.download_base import DownloadJob
+    from invokeai.app.services.model_install.model_install_common import ModelInstallJob
+    from invokeai.app.services.session_processor.session_processor_common import ProgressImage
+    from invokeai.app.services.session_queue.session_queue_common import (
+        BatchStatus,
+        EnqueueBatchResult,
+        SessionQueueItem,
+        SessionQueueStatus,
+    )
+    from invokeai.backend.model_manager.config import AnyModelConfig, SubModelType
 
 
 class EventServiceBase:
-    queue_event: str = "queue_event"
-    bulk_download_event: str = "bulk_download_event"
-    download_event: str = "download_event"
-    model_event: str = "model_event"
-
     """Basic event bus, to have an empty stand-in when not needed"""
 
-    def dispatch(self, event_name: str, payload: Any) -> None:
+    def dispatch(self, event: "EventBase") -> None:
         pass
 
-    def _emit_bulk_download_event(self, event_name: str, payload: dict) -> None:
-        """Bulk download events are emitted to a room with queue_id as the room name"""
-        payload["timestamp"] = get_timestamp()
-        self.dispatch(
-            event_name=EventServiceBase.bulk_download_event,
-            payload={"event": event_name, "data": payload},
-        )
+    # region: Invocation
 
-    def __emit_queue_event(self, event_name: str, payload: dict) -> None:
-        """Queue events are emitted to a room with queue_id as the room name"""
-        payload["timestamp"] = get_timestamp()
-        self.dispatch(
-            event_name=EventServiceBase.queue_event,
-            payload={"event": event_name, "data": payload},
-        )
+    def emit_invocation_started(self, queue_item: "SessionQueueItem", invocation: "BaseInvocation") -> None:
+        """Emitted when an invocation is started"""
+        self.dispatch(InvocationStartedEvent.build(queue_item, invocation))
 
-    def __emit_download_event(self, event_name: str, payload: dict) -> None:
-        payload["timestamp"] = get_timestamp()
-        self.dispatch(
-            event_name=EventServiceBase.download_event,
-            payload={"event": event_name, "data": payload},
-        )
-
-    def __emit_model_event(self, event_name: str, payload: dict) -> None:
-        payload["timestamp"] = get_timestamp()
-        self.dispatch(
-            event_name=EventServiceBase.model_event,
-            payload={"event": event_name, "data": payload},
-        )
-
-    # Define events here for every event in the system.
-    # This will make them easier to integrate until we find a schema generator.
-    def emit_generator_progress(
+    def emit_invocation_denoise_progress(
         self,
-        queue_id: str,
-        queue_item_id: int,
-        queue_batch_id: str,
-        graph_execution_state_id: str,
-        node_id: str,
-        source_node_id: str,
-        progress_image: Optional[ProgressImage],
-        step: int,
-        order: int,
-        total_steps: int,
+        queue_item: "SessionQueueItem",
+        invocation: "BaseInvocation",
+        intermediate_state: PipelineIntermediateState,
+        progress_image: "ProgressImage",
     ) -> None:
-        """Emitted when there is generation progress"""
-        self.__emit_queue_event(
-            event_name="generator_progress",
-            payload={
-                "queue_id": queue_id,
-                "queue_item_id": queue_item_id,
-                "queue_batch_id": queue_batch_id,
-                "graph_execution_state_id": graph_execution_state_id,
-                "node_id": node_id,
-                "source_node_id": source_node_id,
-                "progress_image": progress_image.model_dump(mode="json") if progress_image is not None else None,
-                "step": step,
-                "order": order,
-                "total_steps": total_steps,
-            },
-        )
+        """Emitted at each step during denoising of an invocation."""
+        self.dispatch(InvocationDenoiseProgressEvent.build(queue_item, invocation, intermediate_state, progress_image))
 
     def emit_invocation_complete(
-        self,
-        queue_id: str,
-        queue_item_id: int,
-        queue_batch_id: str,
-        graph_execution_state_id: str,
-        result: dict,
-        node: dict,
-        source_node_id: str,
+        self, queue_item: "SessionQueueItem", invocation: "BaseInvocation", output: "BaseInvocationOutput"
     ) -> None:
-        """Emitted when an invocation has completed"""
-        self.__emit_queue_event(
-            event_name="invocation_complete",
-            payload={
-                "queue_id": queue_id,
-                "queue_item_id": queue_item_id,
-                "queue_batch_id": queue_batch_id,
-                "graph_execution_state_id": graph_execution_state_id,
-                "node": node,
-                "source_node_id": source_node_id,
-                "result": result,
-            },
-        )
+        """Emitted when an invocation is complete"""
+        self.dispatch(InvocationCompleteEvent.build(queue_item, invocation, output))
 
     def emit_invocation_error(
         self,
-        queue_id: str,
-        queue_item_id: int,
-        queue_batch_id: str,
-        graph_execution_state_id: str,
-        node: dict,
-        source_node_id: str,
+        queue_item: "SessionQueueItem",
+        invocation: "BaseInvocation",
         error_type: str,
-        error: str,
+        error_message: str,
+        error_traceback: str,
     ) -> None:
-        """Emitted when an invocation has completed"""
-        self.__emit_queue_event(
-            event_name="invocation_error",
-            payload={
-                "queue_id": queue_id,
-                "queue_item_id": queue_item_id,
-                "queue_batch_id": queue_batch_id,
-                "graph_execution_state_id": graph_execution_state_id,
-                "node": node,
-                "source_node_id": source_node_id,
-                "error_type": error_type,
-                "error": error,
-            },
-        )
+        """Emitted when an invocation encounters an error"""
+        self.dispatch(InvocationErrorEvent.build(queue_item, invocation, error_type, error_message, error_traceback))
 
-    def emit_invocation_started(
-        self,
-        queue_id: str,
-        queue_item_id: int,
-        queue_batch_id: str,
-        graph_execution_state_id: str,
-        node: dict,
-        source_node_id: str,
-    ) -> None:
-        """Emitted when an invocation has started"""
-        self.__emit_queue_event(
-            event_name="invocation_started",
-            payload={
-                "queue_id": queue_id,
-                "queue_item_id": queue_item_id,
-                "queue_batch_id": queue_batch_id,
-                "graph_execution_state_id": graph_execution_state_id,
-                "node": node,
-                "source_node_id": source_node_id,
-            },
-        )
+    # endregion
 
-    def emit_graph_execution_complete(
-        self, queue_id: str, queue_item_id: int, queue_batch_id: str, graph_execution_state_id: str
-    ) -> None:
-        """Emitted when a session has completed all invocations"""
-        self.__emit_queue_event(
-            event_name="graph_execution_state_complete",
-            payload={
-                "queue_id": queue_id,
-                "queue_item_id": queue_item_id,
-                "queue_batch_id": queue_batch_id,
-                "graph_execution_state_id": graph_execution_state_id,
-            },
-        )
-
-    def emit_model_load_started(
-        self,
-        queue_id: str,
-        queue_item_id: int,
-        queue_batch_id: str,
-        graph_execution_state_id: str,
-        model_config: AnyModelConfig,
-        submodel_type: Optional[SubModelType] = None,
-    ) -> None:
-        """Emitted when a model is requested"""
-        self.__emit_queue_event(
-            event_name="model_load_started",
-            payload={
-                "queue_id": queue_id,
-                "queue_item_id": queue_item_id,
-                "queue_batch_id": queue_batch_id,
-                "graph_execution_state_id": graph_execution_state_id,
-                "model_config": model_config.model_dump(mode="json"),
-                "submodel_type": submodel_type,
-            },
-        )
-
-    def emit_model_load_completed(
-        self,
-        queue_id: str,
-        queue_item_id: int,
-        queue_batch_id: str,
-        graph_execution_state_id: str,
-        model_config: AnyModelConfig,
-        submodel_type: Optional[SubModelType] = None,
-    ) -> None:
-        """Emitted when a model is correctly loaded (returns model info)"""
-        self.__emit_queue_event(
-            event_name="model_load_completed",
-            payload={
-                "queue_id": queue_id,
-                "queue_item_id": queue_item_id,
-                "queue_batch_id": queue_batch_id,
-                "graph_execution_state_id": graph_execution_state_id,
-                "model_config": model_config.model_dump(mode="json"),
-                "submodel_type": submodel_type,
-            },
-        )
-
-    def emit_session_canceled(
-        self,
-        queue_id: str,
-        queue_item_id: int,
-        queue_batch_id: str,
-        graph_execution_state_id: str,
-    ) -> None:
-        """Emitted when a session is canceled"""
-        self.__emit_queue_event(
-            event_name="session_canceled",
-            payload={
-                "queue_id": queue_id,
-                "queue_item_id": queue_item_id,
-                "queue_batch_id": queue_batch_id,
-                "graph_execution_state_id": graph_execution_state_id,
-            },
-        )
+    # region Queue
 
     def emit_queue_item_status_changed(
-        self,
-        session_queue_item: SessionQueueItem,
-        batch_status: BatchStatus,
-        queue_status: SessionQueueStatus,
+        self, queue_item: "SessionQueueItem", batch_status: "BatchStatus", queue_status: "SessionQueueStatus"
     ) -> None:
         """Emitted when a queue item's status changes"""
-        self.__emit_queue_event(
-            event_name="queue_item_status_changed",
-            payload={
-                "queue_id": queue_status.queue_id,
-                "queue_item": {
-                    "queue_id": session_queue_item.queue_id,
-                    "item_id": session_queue_item.item_id,
-                    "status": session_queue_item.status,
-                    "batch_id": session_queue_item.batch_id,
-                    "session_id": session_queue_item.session_id,
-                    "error": session_queue_item.error,
-                    "created_at": str(session_queue_item.created_at) if session_queue_item.created_at else None,
-                    "updated_at": str(session_queue_item.updated_at) if session_queue_item.updated_at else None,
-                    "started_at": str(session_queue_item.started_at) if session_queue_item.started_at else None,
-                    "completed_at": str(session_queue_item.completed_at) if session_queue_item.completed_at else None,
-                },
-                "batch_status": batch_status.model_dump(mode="json"),
-                "queue_status": queue_status.model_dump(mode="json"),
-            },
-        )
+        self.dispatch(QueueItemStatusChangedEvent.build(queue_item, batch_status, queue_status))
 
-    def emit_batch_enqueued(self, enqueue_result: EnqueueBatchResult) -> None:
+    def emit_batch_enqueued(self, enqueue_result: "EnqueueBatchResult") -> None:
         """Emitted when a batch is enqueued"""
-        self.__emit_queue_event(
-            event_name="batch_enqueued",
-            payload={
-                "queue_id": enqueue_result.queue_id,
-                "batch_id": enqueue_result.batch.batch_id,
-                "enqueued": enqueue_result.enqueued,
-            },
-        )
+        self.dispatch(BatchEnqueuedEvent.build(enqueue_result))
 
     def emit_queue_cleared(self, queue_id: str) -> None:
-        """Emitted when the queue is cleared"""
-        self.__emit_queue_event(
-            event_name="queue_cleared",
-            payload={"queue_id": queue_id},
-        )
+        """Emitted when a queue is cleared"""
+        self.dispatch(QueueClearedEvent.build(queue_id))
 
-    def emit_download_started(self, source: str, download_path: str) -> None:
-        """
-        Emit when a download job is started.
+    # endregion
 
-        :param url: The downloaded url
-        """
-        self.__emit_download_event(
-            event_name="download_started",
-            payload={"source": source, "download_path": download_path},
-        )
+    # region Download
 
-    def emit_download_progress(self, source: str, download_path: str, current_bytes: int, total_bytes: int) -> None:
-        """
-        Emit "download_progress" events at regular intervals during a download job.
+    def emit_download_started(self, job: "DownloadJob") -> None:
+        """Emitted when a download is started"""
+        self.dispatch(DownloadStartedEvent.build(job))
 
-        :param source: The downloaded source
-        :param download_path: The local downloaded file
-        :param current_bytes: Number of bytes downloaded so far
-        :param total_bytes: The size of the file being downloaded (if known)
-        """
-        self.__emit_download_event(
-            event_name="download_progress",
-            payload={
-                "source": source,
-                "download_path": download_path,
-                "current_bytes": current_bytes,
-                "total_bytes": total_bytes,
-            },
-        )
+    def emit_download_progress(self, job: "DownloadJob") -> None:
+        """Emitted at intervals during a download"""
+        self.dispatch(DownloadProgressEvent.build(job))
 
-    def emit_download_complete(self, source: str, download_path: str, total_bytes: int) -> None:
-        """
-        Emit a "download_complete" event at the end of a successful download.
+    def emit_download_complete(self, job: "DownloadJob") -> None:
+        """Emitted when a download is completed"""
+        self.dispatch(DownloadCompleteEvent.build(job))
 
-        :param source: Source URL
-        :param download_path: Path to the locally downloaded file
-        :param total_bytes: The size of the downloaded file
-        """
-        self.__emit_download_event(
-            event_name="download_complete",
-            payload={
-                "source": source,
-                "download_path": download_path,
-                "total_bytes": total_bytes,
-            },
-        )
+    def emit_download_cancelled(self, job: "DownloadJob") -> None:
+        """Emitted when a download is cancelled"""
+        self.dispatch(DownloadCancelledEvent.build(job))
 
-    def emit_download_cancelled(self, source: str) -> None:
-        """Emit a "download_cancelled" event in the event that the download was cancelled by user."""
-        self.__emit_download_event(
-            event_name="download_cancelled",
-            payload={
-                "source": source,
-            },
-        )
+    def emit_download_error(self, job: "DownloadJob") -> None:
+        """Emitted when a download encounters an error"""
+        self.dispatch(DownloadErrorEvent.build(job))
 
-    def emit_download_error(self, source: str, error_type: str, error: str) -> None:
-        """
-        Emit a "download_error" event when an download job encounters an exception.
+    # endregion
 
-        :param source: Source URL
-        :param error_type: The name of the exception that raised the error
-        :param error: The traceback from this error
-        """
-        self.__emit_download_event(
-            event_name="download_error",
-            payload={
-                "source": source,
-                "error_type": error_type,
-                "error": error,
-            },
-        )
+    # region Model loading
 
-    def emit_model_install_downloading(
-        self,
-        source: str,
-        local_path: str,
-        bytes: int,
-        total_bytes: int,
-        parts: List[Dict[str, Union[str, int]]],
-        id: int,
+    def emit_model_load_started(self, config: "AnyModelConfig", submodel_type: Optional["SubModelType"] = None) -> None:
+        """Emitted when a model load is started."""
+        self.dispatch(ModelLoadStartedEvent.build(config, submodel_type))
+
+    def emit_model_load_complete(
+        self, config: "AnyModelConfig", submodel_type: Optional["SubModelType"] = None
     ) -> None:
-        """
-        Emit at intervals while the install job is in progress (remote models only).
+        """Emitted when a model load is complete."""
+        self.dispatch(ModelLoadCompleteEvent.build(config, submodel_type))
 
-        :param source: Source of the model
-        :param local_path: Where model is downloading to
-        :param parts: Progress of downloading URLs that comprise the model, if any.
-        :param bytes: Number of bytes downloaded so far.
-        :param total_bytes: Total size of download, including all files.
-        This emits a Dict with keys "source", "local_path", "bytes" and "total_bytes".
-        """
-        self.__emit_model_event(
-            event_name="model_install_downloading",
-            payload={
-                "source": source,
-                "local_path": local_path,
-                "bytes": bytes,
-                "total_bytes": total_bytes,
-                "parts": parts,
-                "id": id,
-            },
-        )
+    # endregion
 
-    def emit_model_install_downloads_done(self, source: str) -> None:
-        """
-        Emit once when all parts are downloaded, but before the probing and registration start.
+    # region Model install
 
-        :param source: Source of the model; local path, repo_id or url
-        """
-        self.__emit_model_event(
-            event_name="model_install_downloads_done",
-            payload={"source": source},
-        )
+    def emit_model_install_download_started(self, job: "ModelInstallJob") -> None:
+        """Emitted at intervals while the install job is started (remote models only)."""
+        self.dispatch(ModelInstallDownloadStartedEvent.build(job))
 
-    def emit_model_install_running(self, source: str) -> None:
-        """
-        Emit once when an install job becomes active.
+    def emit_model_install_download_progress(self, job: "ModelInstallJob") -> None:
+        """Emitted at intervals while the install job is in progress (remote models only)."""
+        self.dispatch(ModelInstallDownloadProgressEvent.build(job))
 
-        :param source: Source of the model; local path, repo_id or url
-        """
-        self.__emit_model_event(
-            event_name="model_install_running",
-            payload={"source": source},
-        )
+    def emit_model_install_downloads_complete(self, job: "ModelInstallJob") -> None:
+        self.dispatch(ModelInstallDownloadsCompleteEvent.build(job))
 
-    def emit_model_install_completed(self, source: str, key: str, id: int, total_bytes: Optional[int] = None) -> None:
-        """
-        Emit when an install job is completed successfully.
+    def emit_model_install_started(self, job: "ModelInstallJob") -> None:
+        """Emitted once when an install job is started (after any download)."""
+        self.dispatch(ModelInstallStartedEvent.build(job))
 
-        :param source: Source of the model; local path, repo_id or url
-        :param key: Model config record key
-        :param total_bytes: Size of the model (may be None for installation of a local path)
-        """
-        self.__emit_model_event(
-            event_name="model_install_completed",
-            payload={"source": source, "total_bytes": total_bytes, "key": key, "id": id},
-        )
+    def emit_model_install_complete(self, job: "ModelInstallJob") -> None:
+        """Emitted when an install job is completed successfully."""
+        self.dispatch(ModelInstallCompleteEvent.build(job))
 
-    def emit_model_install_cancelled(self, source: str, id: int) -> None:
-        """
-        Emit when an install job is cancelled.
+    def emit_model_install_cancelled(self, job: "ModelInstallJob") -> None:
+        """Emitted when an install job is cancelled."""
+        self.dispatch(ModelInstallCancelledEvent.build(job))
 
-        :param source: Source of the model; local path, repo_id or url
-        """
-        self.__emit_model_event(
-            event_name="model_install_cancelled",
-            payload={"source": source, "id": id},
-        )
+    def emit_model_install_error(self, job: "ModelInstallJob") -> None:
+        """Emitted when an install job encounters an exception."""
+        self.dispatch(ModelInstallErrorEvent.build(job))
 
-    def emit_model_install_error(self, source: str, error_type: str, error: str, id: int) -> None:
-        """
-        Emit when an install job encounters an exception.
+    # endregion
 
-        :param source: Source of the model
-        :param error_type: The name of the exception
-        :param error: A text description of the exception
-        """
-        self.__emit_model_event(
-            event_name="model_install_error",
-            payload={"source": source, "error_type": error_type, "error": error, "id": id},
-        )
+    # region Bulk image download
 
     def emit_bulk_download_started(
         self, bulk_download_id: str, bulk_download_item_id: str, bulk_download_item_name: str
     ) -> None:
-        """Emitted when a bulk download starts"""
-        self._emit_bulk_download_event(
-            event_name="bulk_download_started",
-            payload={
-                "bulk_download_id": bulk_download_id,
-                "bulk_download_item_id": bulk_download_item_id,
-                "bulk_download_item_name": bulk_download_item_name,
-            },
-        )
+        """Emitted when a bulk image download is started"""
+        self.dispatch(BulkDownloadStartedEvent.build(bulk_download_id, bulk_download_item_id, bulk_download_item_name))
 
-    def emit_bulk_download_completed(
+    def emit_bulk_download_complete(
         self, bulk_download_id: str, bulk_download_item_id: str, bulk_download_item_name: str
     ) -> None:
-        """Emitted when a bulk download completes"""
-        self._emit_bulk_download_event(
-            event_name="bulk_download_completed",
-            payload={
-                "bulk_download_id": bulk_download_id,
-                "bulk_download_item_id": bulk_download_item_id,
-                "bulk_download_item_name": bulk_download_item_name,
-            },
-        )
+        """Emitted when a bulk image download is complete"""
+        self.dispatch(BulkDownloadCompleteEvent.build(bulk_download_id, bulk_download_item_id, bulk_download_item_name))
 
-    def emit_bulk_download_failed(
+    def emit_bulk_download_error(
         self, bulk_download_id: str, bulk_download_item_id: str, bulk_download_item_name: str, error: str
     ) -> None:
-        """Emitted when a bulk download fails"""
-        self._emit_bulk_download_event(
-            event_name="bulk_download_failed",
-            payload={
-                "bulk_download_id": bulk_download_id,
-                "bulk_download_item_id": bulk_download_item_id,
-                "bulk_download_item_name": bulk_download_item_name,
-                "error": error,
-            },
+        """Emitted when a bulk image download has an error"""
+        self.dispatch(
+            BulkDownloadErrorEvent.build(bulk_download_id, bulk_download_item_id, bulk_download_item_name, error)
         )
+
+    # endregion

invokeai/app/services/events/events_common.py

@@ -0,0 +1,628 @@
+from math import floor
+from typing import TYPE_CHECKING, Any, ClassVar, Coroutine, Generic, Optional, Protocol, TypeAlias, TypeVar
+
+from fastapi_events.handlers.local import local_handler
+from fastapi_events.registry.payload_schema import registry as payload_schema
+from pydantic import BaseModel, ConfigDict, Field
+
+from invokeai.app.services.session_processor.session_processor_common import ProgressImage
+from invokeai.app.services.session_queue.session_queue_common import (
+    QUEUE_ITEM_STATUS,
+    BatchStatus,
+    EnqueueBatchResult,
+    SessionQueueItem,
+    SessionQueueStatus,
+)
+from invokeai.app.services.shared.graph import AnyInvocation, AnyInvocationOutput
+from invokeai.app.util.misc import get_timestamp
+from invokeai.backend.model_manager.config import AnyModelConfig, SubModelType
+from invokeai.backend.stable_diffusion.diffusers_pipeline import PipelineIntermediateState
+
+if TYPE_CHECKING:
+    from invokeai.app.services.download.download_base import DownloadJob
+    from invokeai.app.services.model_install.model_install_common import ModelInstallJob
+
+
+class EventBase(BaseModel):
+    """Base class for all events. All events must inherit from this class.
+
+    Events must define a class attribute `__event_name__` to identify the event.
+
+    All other attributes should be defined as normal for a pydantic model.
+
+    A timestamp is automatically added to the event when it is created.
+    """
+
+    __event_name__: ClassVar[str]
+    timestamp: int = Field(description="The timestamp of the event", default_factory=get_timestamp)
+
+    model_config = ConfigDict(json_schema_serialization_defaults_required=True)
+
+    @classmethod
+    def get_events(cls) -> set[type["EventBase"]]:
+        """Get a set of all event models."""
+
+        event_subclasses: set[type["EventBase"]] = set()
+        for subclass in cls.__subclasses__():
+            # We only want to include subclasses that are event models, not intermediary classes
+            if hasattr(subclass, "__event_name__"):
+                event_subclasses.add(subclass)
+            event_subclasses.update(subclass.get_events())
+
+        return event_subclasses
+
+
+TEvent = TypeVar("TEvent", bound=EventBase, contravariant=True)
+
+FastAPIEvent: TypeAlias = tuple[str, TEvent]
+"""
+A tuple representing a `fastapi-events` event, with the event name and payload.
+Provide a generic type to `TEvent` to specify the payload type.
+"""
+
+
+class FastAPIEventFunc(Protocol, Generic[TEvent]):
+    def __call__(self, event: FastAPIEvent[TEvent]) -> Optional[Coroutine[Any, Any, None]]: ...
+
+
+def register_events(events: set[type[TEvent]] | type[TEvent], func: FastAPIEventFunc[TEvent]) -> None:
+    """Register a function to handle specific events.
+
+    :param events: An event or set of events to handle
+    :param func: The function to handle the events
+    """
+    events = events if isinstance(events, set) else {events}
+    for event in events:
+        assert hasattr(event, "__event_name__")
+        local_handler.register(event_name=event.__event_name__, _func=func)  # pyright: ignore [reportUnknownMemberType, reportUnknownArgumentType, reportAttributeAccessIssue]
+
+
+class QueueEventBase(EventBase):
+    """Base class for queue events"""
+
+    queue_id: str = Field(description="The ID of the queue")
+
+
+class QueueItemEventBase(QueueEventBase):
+    """Base class for queue item events"""
+
+    item_id: int = Field(description="The ID of the queue item")
+    batch_id: str = Field(description="The ID of the queue batch")
+
+
+class InvocationEventBase(QueueItemEventBase):
+    """Base class for invocation events"""
+
+    session_id: str = Field(description="The ID of the session (aka graph execution state)")
+    queue_id: str = Field(description="The ID of the queue")
+    item_id: int = Field(description="The ID of the queue item")
+    batch_id: str = Field(description="The ID of the queue batch")
+    session_id: str = Field(description="The ID of the session (aka graph execution state)")
+    invocation: AnyInvocation = Field(description="The ID of the invocation")
+    invocation_source_id: str = Field(description="The ID of the prepared invocation's source node")
+
+
+@payload_schema.register
+class InvocationStartedEvent(InvocationEventBase):
+    """Event model for invocation_started"""
+
+    __event_name__ = "invocation_started"
+
+    @classmethod
+    def build(cls, queue_item: SessionQueueItem, invocation: AnyInvocation) -> "InvocationStartedEvent":
+        return cls(
+            queue_id=queue_item.queue_id,
+            item_id=queue_item.item_id,
+            batch_id=queue_item.batch_id,
+            session_id=queue_item.session_id,
+            invocation=invocation,
+            invocation_source_id=queue_item.session.prepared_source_mapping[invocation.id],
+        )
+
+
+@payload_schema.register
+class InvocationDenoiseProgressEvent(InvocationEventBase):
+    """Event model for invocation_denoise_progress"""
+
+    __event_name__ = "invocation_denoise_progress"
+
+    progress_image: ProgressImage = Field(description="The progress image sent at each step during processing")
+    step: int = Field(description="The current step of the invocation")
+    total_steps: int = Field(description="The total number of steps in the invocation")
+    order: int = Field(description="The order of the invocation in the session")
+    percentage: float = Field(description="The percentage of completion of the invocation")
+
+    @classmethod
+    def build(
+        cls,
+        queue_item: SessionQueueItem,
+        invocation: AnyInvocation,
+        intermediate_state: PipelineIntermediateState,
+        progress_image: ProgressImage,
+    ) -> "InvocationDenoiseProgressEvent":
+        step = intermediate_state.step
+        total_steps = intermediate_state.total_steps
+        order = intermediate_state.order
+        return cls(
+            queue_id=queue_item.queue_id,
+            item_id=queue_item.item_id,
+            batch_id=queue_item.batch_id,
+            session_id=queue_item.session_id,
+            invocation=invocation,
+            invocation_source_id=queue_item.session.prepared_source_mapping[invocation.id],
+            progress_image=progress_image,
+            step=step,
+            total_steps=total_steps,
+            order=order,
+            percentage=cls.calc_percentage(step, total_steps, order),
+        )
+
+    @staticmethod
+    def calc_percentage(step: int, total_steps: int, scheduler_order: float) -> float:
+        """Calculate the percentage of completion of denoising."""
+        if total_steps == 0:
+            return 0.0
+        if scheduler_order == 2:
+            return floor((step + 1 + 1) / 2) / floor((total_steps + 1) / 2)
+        # order == 1
+        return (step + 1 + 1) / (total_steps + 1)
+
+
+@payload_schema.register
+class InvocationCompleteEvent(InvocationEventBase):
+    """Event model for invocation_complete"""
+
+    __event_name__ = "invocation_complete"
+
+    result: AnyInvocationOutput = Field(description="The result of the invocation")
+
+    @classmethod
+    def build(
+        cls, queue_item: SessionQueueItem, invocation: AnyInvocation, result: AnyInvocationOutput
+    ) -> "InvocationCompleteEvent":
+        return cls(
+            queue_id=queue_item.queue_id,
+            item_id=queue_item.item_id,
+            batch_id=queue_item.batch_id,
+            session_id=queue_item.session_id,
+            invocation=invocation,
+            invocation_source_id=queue_item.session.prepared_source_mapping[invocation.id],
+            result=result,
+        )
+
+
+@payload_schema.register
+class InvocationErrorEvent(InvocationEventBase):
+    """Event model for invocation_error"""
+
+    __event_name__ = "invocation_error"
+
+    error_type: str = Field(description="The error type")
+    error_message: str = Field(description="The error message")
+    error_traceback: str = Field(description="The error traceback")
+    user_id: Optional[str] = Field(default=None, description="The ID of the user who created the invocation")
+    project_id: Optional[str] = Field(default=None, description="The ID of the user who created the invocation")
+
+    @classmethod
+    def build(
+        cls,
+        queue_item: SessionQueueItem,
+        invocation: AnyInvocation,
+        error_type: str,
+        error_message: str,
+        error_traceback: str,
+    ) -> "InvocationErrorEvent":
+        return cls(
+            queue_id=queue_item.queue_id,
+            item_id=queue_item.item_id,
+            batch_id=queue_item.batch_id,
+            session_id=queue_item.session_id,
+            invocation=invocation,
+            invocation_source_id=queue_item.session.prepared_source_mapping[invocation.id],
+            error_type=error_type,
+            error_message=error_message,
+            error_traceback=error_traceback,
+            user_id=getattr(queue_item, "user_id", None),
+            project_id=getattr(queue_item, "project_id", None),
+        )
+
+
+@payload_schema.register
+class QueueItemStatusChangedEvent(QueueItemEventBase):
+    """Event model for queue_item_status_changed"""
+
+    __event_name__ = "queue_item_status_changed"
+
+    status: QUEUE_ITEM_STATUS = Field(description="The new status of the queue item")
+    error_type: Optional[str] = Field(default=None, description="The error type, if any")
+    error_message: Optional[str] = Field(default=None, description="The error message, if any")
+    error_traceback: Optional[str] = Field(default=None, description="The error traceback, if any")
+    created_at: Optional[str] = Field(default=None, description="The timestamp when the queue item was created")
+    updated_at: Optional[str] = Field(default=None, description="The timestamp when the queue item was last updated")
+    started_at: Optional[str] = Field(default=None, description="The timestamp when the queue item was started")
+    completed_at: Optional[str] = Field(default=None, description="The timestamp when the queue item was completed")
+    batch_status: BatchStatus = Field(description="The status of the batch")
+    queue_status: SessionQueueStatus = Field(description="The status of the queue")
+    session_id: str = Field(description="The ID of the session (aka graph execution state)")
+
+    @classmethod
+    def build(
+        cls, queue_item: SessionQueueItem, batch_status: BatchStatus, queue_status: SessionQueueStatus
+    ) -> "QueueItemStatusChangedEvent":
+        return cls(
+            queue_id=queue_item.queue_id,
+            item_id=queue_item.item_id,
+            batch_id=queue_item.batch_id,
+            session_id=queue_item.session_id,
+            status=queue_item.status,
+            error_type=queue_item.error_type,
+            error_message=queue_item.error_message,
+            error_traceback=queue_item.error_traceback,
+            created_at=str(queue_item.created_at) if queue_item.created_at else None,
+            updated_at=str(queue_item.updated_at) if queue_item.updated_at else None,
+            started_at=str(queue_item.started_at) if queue_item.started_at else None,
+            completed_at=str(queue_item.completed_at) if queue_item.completed_at else None,
+            batch_status=batch_status,
+            queue_status=queue_status,
+        )
+
+
+@payload_schema.register
+class BatchEnqueuedEvent(QueueEventBase):
+    """Event model for batch_enqueued"""
+
+    __event_name__ = "batch_enqueued"
+
+    batch_id: str = Field(description="The ID of the batch")
+    enqueued: int = Field(description="The number of invocations enqueued")
+    requested: int = Field(
+        description="The number of invocations initially requested to be enqueued (may be less than enqueued if queue was full)"
+    )
+    priority: int = Field(description="The priority of the batch")
+
+    @classmethod
+    def build(cls, enqueue_result: EnqueueBatchResult) -> "BatchEnqueuedEvent":
+        return cls(
+            queue_id=enqueue_result.queue_id,
+            batch_id=enqueue_result.batch.batch_id,
+            enqueued=enqueue_result.enqueued,
+            requested=enqueue_result.requested,
+            priority=enqueue_result.priority,
+        )
+
+
+@payload_schema.register
+class QueueClearedEvent(QueueEventBase):
+    """Event model for queue_cleared"""
+
+    __event_name__ = "queue_cleared"
+
+    @classmethod
+    def build(cls, queue_id: str) -> "QueueClearedEvent":
+        return cls(queue_id=queue_id)
+
+
+class DownloadEventBase(EventBase):
+    """Base class for events associated with a download"""
+
+    source: str = Field(description="The source of the download")
+
+
+@payload_schema.register
+class DownloadStartedEvent(DownloadEventBase):
+    """Event model for download_started"""
+
+    __event_name__ = "download_started"
+
+    download_path: str = Field(description="The local path where the download is saved")
+
+    @classmethod
+    def build(cls, job: "DownloadJob") -> "DownloadStartedEvent":
+        assert job.download_path
+        return cls(source=str(job.source), download_path=job.download_path.as_posix())
+
+
+@payload_schema.register
+class DownloadProgressEvent(DownloadEventBase):
+    """Event model for download_progress"""
+
+    __event_name__ = "download_progress"
+
+    download_path: str = Field(description="The local path where the download is saved")
+    current_bytes: int = Field(description="The number of bytes downloaded so far")
+    total_bytes: int = Field(description="The total number of bytes to be downloaded")
+
+    @classmethod
+    def build(cls, job: "DownloadJob") -> "DownloadProgressEvent":
+        assert job.download_path
+        return cls(
+            source=str(job.source),
+            download_path=job.download_path.as_posix(),
+            current_bytes=job.bytes,
+            total_bytes=job.total_bytes,
+        )
+
+
+@payload_schema.register
+class DownloadCompleteEvent(DownloadEventBase):
+    """Event model for download_complete"""
+
+    __event_name__ = "download_complete"
+
+    download_path: str = Field(description="The local path where the download is saved")
+    total_bytes: int = Field(description="The total number of bytes downloaded")
+
+    @classmethod
+    def build(cls, job: "DownloadJob") -> "DownloadCompleteEvent":
+        assert job.download_path
+        return cls(source=str(job.source), download_path=job.download_path.as_posix(), total_bytes=job.total_bytes)
+
+
+@payload_schema.register
+class DownloadCancelledEvent(DownloadEventBase):
+    """Event model for download_cancelled"""
+
+    __event_name__ = "download_cancelled"
+
+    @classmethod
+    def build(cls, job: "DownloadJob") -> "DownloadCancelledEvent":
+        return cls(source=str(job.source))
+
+
+@payload_schema.register
+class DownloadErrorEvent(DownloadEventBase):
+    """Event model for download_error"""
+
+    __event_name__ = "download_error"
+
+    error_type: str = Field(description="The type of error")
+    error: str = Field(description="The error message")
+
+    @classmethod
+    def build(cls, job: "DownloadJob") -> "DownloadErrorEvent":
+        assert job.error_type
+        assert job.error
+        return cls(source=str(job.source), error_type=job.error_type, error=job.error)
+
+
+class ModelEventBase(EventBase):
+    """Base class for events associated with a model"""
+
+
+@payload_schema.register
+class ModelLoadStartedEvent(ModelEventBase):
+    """Event model for model_load_started"""
+
+    __event_name__ = "model_load_started"
+
+    config: AnyModelConfig = Field(description="The model's config")
+    submodel_type: Optional[SubModelType] = Field(default=None, description="The submodel type, if any")
+
+    @classmethod
+    def build(cls, config: AnyModelConfig, submodel_type: Optional[SubModelType] = None) -> "ModelLoadStartedEvent":
+        return cls(config=config, submodel_type=submodel_type)
+
+
+@payload_schema.register
+class ModelLoadCompleteEvent(ModelEventBase):
+    """Event model for model_load_complete"""
+
+    __event_name__ = "model_load_complete"
+
+    config: AnyModelConfig = Field(description="The model's config")
+    submodel_type: Optional[SubModelType] = Field(default=None, description="The submodel type, if any")
+
+    @classmethod
+    def build(cls, config: AnyModelConfig, submodel_type: Optional[SubModelType] = None) -> "ModelLoadCompleteEvent":
+        return cls(config=config, submodel_type=submodel_type)
+
+
+@payload_schema.register
+class ModelInstallDownloadStartedEvent(ModelEventBase):
+    """Event model for model_install_download_started"""
+
+    __event_name__ = "model_install_download_started"
+
+    id: int = Field(description="The ID of the install job")
+    source: str = Field(description="Source of the model; local path, repo_id or url")
+    local_path: str = Field(description="Where model is downloading to")
+    bytes: int = Field(description="Number of bytes downloaded so far")
+    total_bytes: int = Field(description="Total size of download, including all files")
+    parts: list[dict[str, int | str]] = Field(
+        description="Progress of downloading URLs that comprise the model, if any"
+    )
+
+    @classmethod
+    def build(cls, job: "ModelInstallJob") -> "ModelInstallDownloadStartedEvent":
+        parts: list[dict[str, str | int]] = [
+            {
+                "url": str(x.source),
+                "local_path": str(x.download_path),
+                "bytes": x.bytes,
+                "total_bytes": x.total_bytes,
+            }
+            for x in job.download_parts
+        ]
+        return cls(
+            id=job.id,
+            source=str(job.source),
+            local_path=job.local_path.as_posix(),
+            parts=parts,
+            bytes=job.bytes,
+            total_bytes=job.total_bytes,
+        )
+
+
+@payload_schema.register
+class ModelInstallDownloadProgressEvent(ModelEventBase):
+    """Event model for model_install_download_progress"""
+
+    __event_name__ = "model_install_download_progress"
+
+    id: int = Field(description="The ID of the install job")
+    source: str = Field(description="Source of the model; local path, repo_id or url")
+    local_path: str = Field(description="Where model is downloading to")
+    bytes: int = Field(description="Number of bytes downloaded so far")
+    total_bytes: int = Field(description="Total size of download, including all files")
+    parts: list[dict[str, int | str]] = Field(
+        description="Progress of downloading URLs that comprise the model, if any"
+    )
+
+    @classmethod
+    def build(cls, job: "ModelInstallJob") -> "ModelInstallDownloadProgressEvent":
+        parts: list[dict[str, str | int]] = [
+            {
+                "url": str(x.source),
+                "local_path": str(x.download_path),
+                "bytes": x.bytes,
+                "total_bytes": x.total_bytes,
+            }
+            for x in job.download_parts
+        ]
+        return cls(
+            id=job.id,
+            source=str(job.source),
+            local_path=job.local_path.as_posix(),
+            parts=parts,
+            bytes=job.bytes,
+            total_bytes=job.total_bytes,
+        )
+
+
+@payload_schema.register
+class ModelInstallDownloadsCompleteEvent(ModelEventBase):
+    """Emitted once when an install job becomes active."""
+
+    __event_name__ = "model_install_downloads_complete"
+
+    id: int = Field(description="The ID of the install job")
+    source: str = Field(description="Source of the model; local path, repo_id or url")
+
+    @classmethod
+    def build(cls, job: "ModelInstallJob") -> "ModelInstallDownloadsCompleteEvent":
+        return cls(id=job.id, source=str(job.source))
+
+
+@payload_schema.register
+class ModelInstallStartedEvent(ModelEventBase):
+    """Event model for model_install_started"""
+
+    __event_name__ = "model_install_started"
+
+    id: int = Field(description="The ID of the install job")
+    source: str = Field(description="Source of the model; local path, repo_id or url")
+
+    @classmethod
+    def build(cls, job: "ModelInstallJob") -> "ModelInstallStartedEvent":
+        return cls(id=job.id, source=str(job.source))
+
+
+@payload_schema.register
+class ModelInstallCompleteEvent(ModelEventBase):
+    """Event model for model_install_complete"""
+
+    __event_name__ = "model_install_complete"
+
+    id: int = Field(description="The ID of the install job")
+    source: str = Field(description="Source of the model; local path, repo_id or url")
+    key: str = Field(description="Model config record key")
+    total_bytes: Optional[int] = Field(description="Size of the model (may be None for installation of a local path)")
+
+    @classmethod
+    def build(cls, job: "ModelInstallJob") -> "ModelInstallCompleteEvent":
+        assert job.config_out is not None
+        return cls(id=job.id, source=str(job.source), key=(job.config_out.key), total_bytes=job.total_bytes)
+
+
+@payload_schema.register
+class ModelInstallCancelledEvent(ModelEventBase):
+    """Event model for model_install_cancelled"""
+
+    __event_name__ = "model_install_cancelled"
+
+    id: int = Field(description="The ID of the install job")
+    source: str = Field(description="Source of the model; local path, repo_id or url")
+
+    @classmethod
+    def build(cls, job: "ModelInstallJob") -> "ModelInstallCancelledEvent":
+        return cls(id=job.id, source=str(job.source))
+
+
+@payload_schema.register
+class ModelInstallErrorEvent(ModelEventBase):
+    """Event model for model_install_error"""
+
+    __event_name__ = "model_install_error"
+
+    id: int = Field(description="The ID of the install job")
+    source: str = Field(description="Source of the model; local path, repo_id or url")
+    error_type: str = Field(description="The name of the exception")
+    error: str = Field(description="A text description of the exception")
+
+    @classmethod
+    def build(cls, job: "ModelInstallJob") -> "ModelInstallErrorEvent":
+        assert job.error_type is not None
+        assert job.error is not None
+        return cls(id=job.id, source=str(job.source), error_type=job.error_type, error=job.error)
+
+
+class BulkDownloadEventBase(EventBase):
+    """Base class for events associated with a bulk image download"""
+
+    bulk_download_id: str = Field(description="The ID of the bulk image download")
+    bulk_download_item_id: str = Field(description="The ID of the bulk image download item")
+    bulk_download_item_name: str = Field(description="The name of the bulk image download item")
+
+
+@payload_schema.register
+class BulkDownloadStartedEvent(BulkDownloadEventBase):
+    """Event model for bulk_download_started"""
+
+    __event_name__ = "bulk_download_started"
+
+    @classmethod
+    def build(
+        cls, bulk_download_id: str, bulk_download_item_id: str, bulk_download_item_name: str
+    ) -> "BulkDownloadStartedEvent":
+        return cls(
+            bulk_download_id=bulk_download_id,
+            bulk_download_item_id=bulk_download_item_id,
+            bulk_download_item_name=bulk_download_item_name,
+        )
+
+
+@payload_schema.register
+class BulkDownloadCompleteEvent(BulkDownloadEventBase):
+    """Event model for bulk_download_complete"""
+
+    __event_name__ = "bulk_download_complete"
+
+    @classmethod
+    def build(
+        cls, bulk_download_id: str, bulk_download_item_id: str, bulk_download_item_name: str
+    ) -> "BulkDownloadCompleteEvent":
+        return cls(
+            bulk_download_id=bulk_download_id,
+            bulk_download_item_id=bulk_download_item_id,
+            bulk_download_item_name=bulk_download_item_name,
+        )
+
+
+@payload_schema.register
+class BulkDownloadErrorEvent(BulkDownloadEventBase):
+    """Event model for bulk_download_error"""
+
+    __event_name__ = "bulk_download_error"
+
+    error: str = Field(description="The error message")
+
+    @classmethod
+    def build(
+        cls, bulk_download_id: str, bulk_download_item_id: str, bulk_download_item_name: str, error: str
+    ) -> "BulkDownloadErrorEvent":
+        return cls(
+            bulk_download_id=bulk_download_id,
+            bulk_download_item_id=bulk_download_item_id,
+            bulk_download_item_name=bulk_download_item_name,
+            error=error,
+        )

invokeai/app/services/events/events_fastapievents.py

@@ -0,0 +1,47 @@
+# Copyright (c) 2022 Kyle Schouviller (https://github.com/kyle0654)
+
+import asyncio
+import threading
+from queue import Empty, Queue
+
+from fastapi_events.dispatcher import dispatch
+
+from invokeai.app.services.events.events_common import (
+    EventBase,
+)
+
+from .events_base import EventServiceBase
+
+
+class FastAPIEventService(EventServiceBase):
+    def __init__(self, event_handler_id: int) -> None:
+        self.event_handler_id = event_handler_id
+        self._queue = Queue[EventBase | None]()
+        self._stop_event = threading.Event()
+        asyncio.create_task(self._dispatch_from_queue(stop_event=self._stop_event))
+
+        super().__init__()
+
+    def stop(self, *args, **kwargs):
+        self._stop_event.set()
+        self._queue.put(None)
+
+    def dispatch(self, event: EventBase) -> None:
+        self._queue.put(event)
+
+    async def _dispatch_from_queue(self, stop_event: threading.Event):
+        """Get events on from the queue and dispatch them, from the correct thread"""
+        while not stop_event.is_set():
+            try:
+                event = self._queue.get(block=False)
+                if not event:  # Probably stopping
+                    continue
+                # Leave the payloads as live pydantic models
+                dispatch(event, middleware_id=self.event_handler_id, payload_schema_dump=False)
+
+            except Empty:
+                await asyncio.sleep(0.1)
+                pass
+
+            except asyncio.CancelledError as e:
+                raise e  # Raise a proper error

invokeai/app/services/image_files/image_files_base.py

@@ -4,9 +4,6 @@ from typing import Optional
 
 from PIL.Image import Image as PILImageType
 
-from invokeai.app.invocations.fields import MetadataField
-from invokeai.app.services.workflow_records.workflow_records_common import WorkflowWithoutID
-
 
 class ImageFileStorageBase(ABC):
     """Low-level service responsible for storing and retrieving image files."""
@@ -33,8 +30,9 @@ class ImageFileStorageBase(ABC):
         self,
         image: PILImageType,
         image_name: str,
-        metadata: Optional[MetadataField] = None,
-        workflow: Optional[WorkflowWithoutID] = None,
+        metadata: Optional[str] = None,
+        workflow: Optional[str] = None,
+        graph: Optional[str] = None,
         thumbnail_size: int = 256,
     ) -> None:
         """Saves an image and a 256x256 WEBP thumbnail. Returns a tuple of the image name, thumbnail name, and created timestamp."""
@@ -46,6 +44,11 @@ class ImageFileStorageBase(ABC):
         pass
 
     @abstractmethod
-    def get_workflow(self, image_name: str) -> Optional[WorkflowWithoutID]:
+    def get_workflow(self, image_name: str) -> Optional[str]:
         """Gets the workflow of an image."""
         pass
+
+    @abstractmethod
+    def get_graph(self, image_name: str) -> Optional[str]:
+        """Gets the graph of an image."""
+        pass

invokeai/app/services/image_files/image_files_disk.py

@@ -7,9 +7,7 @@ from PIL import Image, PngImagePlugin
 from PIL.Image import Image as PILImageType
 from send2trash import send2trash
 
-from invokeai.app.invocations.fields import MetadataField
 from invokeai.app.services.invoker import Invoker
-from invokeai.app.services.workflow_records.workflow_records_common import WorkflowWithoutID
 from invokeai.app.util.thumbnails import get_thumbnail_name, make_thumbnail
 
 from .image_files_base import ImageFileStorageBase
@@ -56,8 +54,9 @@ class DiskImageFileStorage(ImageFileStorageBase):
         self,
         image: PILImageType,
         image_name: str,
-        metadata: Optional[MetadataField] = None,
-        workflow: Optional[WorkflowWithoutID] = None,
+        metadata: Optional[str] = None,
+        workflow: Optional[str] = None,
+        graph: Optional[str] = None,
         thumbnail_size: int = 256,
     ) -> None:
         try:
@@ -68,13 +67,14 @@ class DiskImageFileStorage(ImageFileStorageBase):
             info_dict = {}
 
             if metadata is not None:
-                metadata_json = metadata.model_dump_json()
-                info_dict["invokeai_metadata"] = metadata_json
-                pnginfo.add_text("invokeai_metadata", metadata_json)
+                info_dict["invokeai_metadata"] = metadata
+                pnginfo.add_text("invokeai_metadata", metadata)
             if workflow is not None:
-                workflow_json = workflow.model_dump_json()
-                info_dict["invokeai_workflow"] = workflow_json
-                pnginfo.add_text("invokeai_workflow", workflow_json)
+                info_dict["invokeai_workflow"] = workflow
+                pnginfo.add_text("invokeai_workflow", workflow)
+            if graph is not None:
+                info_dict["invokeai_graph"] = graph
+                pnginfo.add_text("invokeai_graph", graph)
 
             # When saving the image, the image object's info field is not populated. We need to set it
             image.info = info_dict
@@ -129,11 +129,18 @@ class DiskImageFileStorage(ImageFileStorageBase):
         path = path if isinstance(path, Path) else Path(path)
         return path.exists()
 
-    def get_workflow(self, image_name: str) -> WorkflowWithoutID | None:
+    def get_workflow(self, image_name: str) -> str | None:
         image = self.get(image_name)
         workflow = image.info.get("invokeai_workflow", None)
-        if workflow is not None:
-            return WorkflowWithoutID.model_validate_json(workflow)
+        if isinstance(workflow, str):
+            return workflow
+        return None
+
+    def get_graph(self, image_name: str) -> str | None:
+        image = self.get(image_name)
+        graph = image.info.get("invokeai_graph", None)
+        if isinstance(graph, str):
+            return graph
         return None
 
     def __validate_storage_folders(self) -> None:

invokeai/app/services/image_records/image_records_base.py

@@ -80,7 +80,7 @@ class ImageRecordStorageBase(ABC):
         starred: Optional[bool] = False,
         session_id: Optional[str] = None,
         node_id: Optional[str] = None,
-        metadata: Optional[MetadataField] = None,
+        metadata: Optional[str] = None,
     ) -> datetime:
         """Saves an image record."""
         pass

invokeai/app/services/image_records/image_records_sqlite.py

@@ -328,10 +328,9 @@ class SqliteImageRecordStorage(ImageRecordStorageBase):
         starred: Optional[bool] = False,
         session_id: Optional[str] = None,
         node_id: Optional[str] = None,
-        metadata: Optional[MetadataField] = None,
+        metadata: Optional[str] = None,
     ) -> datetime:
         try:
-            metadata_json = metadata.model_dump_json() if metadata is not None else None
             self._lock.acquire()
             self._cursor.execute(
                 """--sql
@@ -358,7 +357,7 @@ class SqliteImageRecordStorage(ImageRecordStorageBase):
                     height,
                     node_id,
                     session_id,
-                    metadata_json,
+                    metadata,
                     is_intermediate,
                     starred,
                     has_workflow,

invokeai/app/services/images/images_base.py

@@ -12,7 +12,6 @@ from invokeai.app.services.image_records.image_records_common import (
 )
 from invokeai.app.services.images.images_common import ImageDTO
 from invokeai.app.services.shared.pagination import OffsetPaginatedResults
-from invokeai.app.services.workflow_records.workflow_records_common import WorkflowWithoutID
 
 
 class ImageServiceABC(ABC):
@@ -51,8 +50,9 @@ class ImageServiceABC(ABC):
         session_id: Optional[str] = None,
         board_id: Optional[str] = None,
         is_intermediate: Optional[bool] = False,
-        metadata: Optional[MetadataField] = None,
-        workflow: Optional[WorkflowWithoutID] = None,
+        metadata: Optional[str] = None,
+        workflow: Optional[str] = None,
+        graph: Optional[str] = None,
     ) -> ImageDTO:
         """Creates an image, storing the file and its metadata."""
         pass
@@ -87,7 +87,12 @@ class ImageServiceABC(ABC):
         pass
 
     @abstractmethod
-    def get_workflow(self, image_name: str) -> Optional[WorkflowWithoutID]:
+    def get_workflow(self, image_name: str) -> Optional[str]:
+        """Gets an image's workflow."""
+        pass
+
+    @abstractmethod
+    def get_graph(self, image_name: str) -> Optional[str]:
         """Gets an image's workflow."""
         pass
 

invokeai/app/services/images/images_default.py

@@ -5,7 +5,6 @@ from PIL.Image import Image as PILImageType
 from invokeai.app.invocations.fields import MetadataField
 from invokeai.app.services.invoker import Invoker
 from invokeai.app.services.shared.pagination import OffsetPaginatedResults
-from invokeai.app.services.workflow_records.workflow_records_common import WorkflowWithoutID
 
 from ..image_files.image_files_common import (
     ImageFileDeleteException,
@@ -42,8 +41,9 @@ class ImageService(ImageServiceABC):
         session_id: Optional[str] = None,
         board_id: Optional[str] = None,
         is_intermediate: Optional[bool] = False,
-        metadata: Optional[MetadataField] = None,
-        workflow: Optional[WorkflowWithoutID] = None,
+        metadata: Optional[str] = None,
+        workflow: Optional[str] = None,
+        graph: Optional[str] = None,
     ) -> ImageDTO:
         if image_origin not in ResourceOrigin:
             raise InvalidOriginException
@@ -64,7 +64,7 @@ class ImageService(ImageServiceABC):
                 image_category=image_category,
                 width=width,
                 height=height,
-                has_workflow=workflow is not None,
+                has_workflow=workflow is not None or graph is not None,
                 # Meta fields
                 is_intermediate=is_intermediate,
                 # Nullable fields
@@ -75,7 +75,7 @@ class ImageService(ImageServiceABC):
             if board_id is not None:
                 self.__invoker.services.board_image_records.add_image_to_board(board_id=board_id, image_name=image_name)
             self.__invoker.services.image_files.save(
-                image_name=image_name, image=image, metadata=metadata, workflow=workflow
+                image_name=image_name, image=image, metadata=metadata, workflow=workflow, graph=graph
             )
             image_dto = self.get_dto(image_name)
 
@@ -157,7 +157,7 @@ class ImageService(ImageServiceABC):
             self.__invoker.services.logger.error("Problem getting image metadata")
             raise e
 
-    def get_workflow(self, image_name: str) -> Optional[WorkflowWithoutID]:
+    def get_workflow(self, image_name: str) -> Optional[str]:
         try:
             return self.__invoker.services.image_files.get_workflow(image_name)
         except ImageFileNotFoundException:
@@ -167,6 +167,16 @@ class ImageService(ImageServiceABC):
             self.__invoker.services.logger.error("Problem getting image workflow")
             raise
 
+    def get_graph(self, image_name: str) -> Optional[str]:
+        try:
+            return self.__invoker.services.image_files.get_graph(image_name)
+        except ImageFileNotFoundException:
+            self.__invoker.services.logger.error("Image file not found")
+            raise
+        except Exception:
+            self.__invoker.services.logger.error("Problem getting image graph")
+            raise
+
     def get_path(self, image_name: str, thumbnail: bool = False) -> str:
         try:
             return str(self.__invoker.services.image_files.get_path(image_name, thumbnail))

invokeai/app/services/invocation_services.py

@@ -53,11 +53,11 @@ class InvocationServices:
         model_images: "ModelImageFileStorageBase",
         model_manager: "ModelManagerServiceBase",
         download_queue: "DownloadQueueServiceBase",
-        performance_statistics: "InvocationStatsServiceBase",
         session_queue: "SessionQueueBase",
         session_processor: "SessionProcessorBase",
         invocation_cache: "InvocationCacheBase",
         names: "NameServiceBase",
+        performance_statistics: "InvocationStatsServiceBase",
         urls: "UrlServiceBase",
         workflow_records: "WorkflowRecordsStorageBase",
         tensors: "ObjectSerializerBase[torch.Tensor]",
@@ -77,11 +77,11 @@ class InvocationServices:
         self.model_images = model_images
         self.model_manager = model_manager
         self.download_queue = download_queue
-        self.performance_statistics = performance_statistics
         self.session_queue = session_queue
         self.session_processor = session_processor
         self.invocation_cache = invocation_cache
         self.names = names
+        self.performance_statistics = performance_statistics
         self.urls = urls
         self.workflow_records = workflow_records
         self.tensors = tensors

invokeai/app/services/invocation_stats/invocation_stats_default.py

@@ -74,9 +74,9 @@ class InvocationStatsService(InvocationStatsServiceBase):
             )
             self._stats[graph_execution_state_id].add_node_execution_stats(node_stats)
 
-    def reset_stats(self):
-        self._stats = {}
-        self._cache_stats = {}
+    def reset_stats(self, graph_execution_state_id: str):
+        self._stats.pop(graph_execution_state_id)
+        self._cache_stats.pop(graph_execution_state_id)
 
     def get_stats(self, graph_execution_state_id: str) -> InvocationStatsSummary:
         graph_stats_summary = self._get_graph_summary(graph_execution_state_id)

invokeai/app/services/model_install/__init__.py

@@ -1,11 +1,13 @@
 """Initialization file for model install service package."""
 
 from .model_install_base import (
+    ModelInstallServiceBase,
+)
+from .model_install_common import (
     HFModelSource,
     InstallStatus,
     LocalModelSource,
     ModelInstallJob,
-    ModelInstallServiceBase,
     ModelSource,
     UnknownInstallJobException,
     URLModelSource,

invokeai/app/services/model_install/model_install_base.py

@@ -1,244 +1,19 @@
 # Copyright 2023 Lincoln D. Stein and the InvokeAI development team
 """Baseclass definitions for the model installer."""
 
-import re
-import traceback
 from abc import ABC, abstractmethod
-from enum import Enum
 from pathlib import Path
-from typing import Any, Dict, List, Literal, Optional, Set, Union
+from typing import Any, Dict, List, Optional, Union
 
-from pydantic import BaseModel, Field, PrivateAttr, field_validator
 from pydantic.networks import AnyHttpUrl
-from typing_extensions import Annotated
 
 from invokeai.app.services.config import InvokeAIAppConfig
-from invokeai.app.services.download import DownloadJob, DownloadQueueServiceBase
+from invokeai.app.services.download import DownloadQueueServiceBase
 from invokeai.app.services.events.events_base import EventServiceBase
 from invokeai.app.services.invoker import Invoker
+from invokeai.app.services.model_install.model_install_common import ModelInstallJob, ModelSource
 from invokeai.app.services.model_records import ModelRecordServiceBase
-from invokeai.backend.model_manager import AnyModelConfig, ModelRepoVariant
-from invokeai.backend.model_manager.config import ModelSourceType
-from invokeai.backend.model_manager.metadata import AnyModelRepoMetadata
-
-
-class InstallStatus(str, Enum):
-    """State of an install job running in the background."""
-
-    WAITING = "waiting"  # waiting to be dequeued
-    DOWNLOADING = "downloading"  # downloading of model files in process
-    DOWNLOADS_DONE = "downloads_done"  # downloading done, waiting to run
-    RUNNING = "running"  # being processed
-    COMPLETED = "completed"  # finished running
-    ERROR = "error"  # terminated with an error message
-    CANCELLED = "cancelled"  # terminated with an error message
-
-
-class ModelInstallPart(BaseModel):
-    url: AnyHttpUrl
-    path: Path
-    bytes: int = 0
-    total_bytes: int = 0
-
-
-class UnknownInstallJobException(Exception):
-    """Raised when the status of an unknown job is requested."""
-
-
-class StringLikeSource(BaseModel):
-    """
-    Base class for model sources, implements functions that lets the source be sorted and indexed.
-
-    These shenanigans let this stuff work:
-
-      source1 = LocalModelSource(path='C:/users/mort/foo.safetensors')
-      mydict = {source1: 'model 1'}
-      assert mydict['C:/users/mort/foo.safetensors'] == 'model 1'
-      assert mydict[LocalModelSource(path='C:/users/mort/foo.safetensors')] == 'model 1'
-
-      source2 = LocalModelSource(path=Path('C:/users/mort/foo.safetensors'))
-      assert source1 == source2
-      assert source1 == 'C:/users/mort/foo.safetensors'
-    """
-
-    def __hash__(self) -> int:
-        """Return hash of the path field, for indexing."""
-        return hash(str(self))
-
-    def __lt__(self, other: object) -> int:
-        """Return comparison of the stringified version, for sorting."""
-        return str(self) < str(other)
-
-    def __eq__(self, other: object) -> bool:
-        """Return equality on the stringified version."""
-        if isinstance(other, Path):
-            return str(self) == other.as_posix()
-        else:
-            return str(self) == str(other)
-
-
-class LocalModelSource(StringLikeSource):
-    """A local file or directory path."""
-
-    path: str | Path
-    inplace: Optional[bool] = False
-    type: Literal["local"] = "local"
-
-    # these methods allow the source to be used in a string-like way,
-    # for example as an index into a dict
-    def __str__(self) -> str:
-        """Return string version of path when string rep needed."""
-        return Path(self.path).as_posix()
-
-
-class HFModelSource(StringLikeSource):
-    """
-    A HuggingFace repo_id with optional variant, sub-folder and access token.
-    Note that the variant option, if not provided to the constructor, will default to fp16, which is
-    what people (almost) always want.
-    """
-
-    repo_id: str
-    variant: Optional[ModelRepoVariant] = ModelRepoVariant.FP16
-    subfolder: Optional[Path] = None
-    access_token: Optional[str] = None
-    type: Literal["hf"] = "hf"
-
-    @field_validator("repo_id")
-    @classmethod
-    def proper_repo_id(cls, v: str) -> str:  # noqa D102
-        if not re.match(r"^([.\w-]+/[.\w-]+)$", v):
-            raise ValueError(f"{v}: invalid repo_id format")
-        return v
-
-    def __str__(self) -> str:
-        """Return string version of repoid when string rep needed."""
-        base: str = self.repo_id
-        if self.variant:
-            base += f":{self.variant or ''}"
-        if self.subfolder:
-            base += f":{self.subfolder}"
-        return base
-
-
-class URLModelSource(StringLikeSource):
-    """A generic URL point to a checkpoint file."""
-
-    url: AnyHttpUrl
-    access_token: Optional[str] = None
-    type: Literal["url"] = "url"
-
-    def __str__(self) -> str:
-        """Return string version of the url when string rep needed."""
-        return str(self.url)
-
-
-ModelSource = Annotated[Union[LocalModelSource, HFModelSource, URLModelSource], Field(discriminator="type")]
-
-MODEL_SOURCE_TO_TYPE_MAP = {
-    URLModelSource: ModelSourceType.Url,
-    HFModelSource: ModelSourceType.HFRepoID,
-    LocalModelSource: ModelSourceType.Path,
-}
-
-
-class ModelInstallJob(BaseModel):
-    """Object that tracks the current status of an install request."""
-
-    id: int = Field(description="Unique ID for this job")
-    status: InstallStatus = Field(default=InstallStatus.WAITING, description="Current status of install process")
-    error_reason: Optional[str] = Field(default=None, description="Information about why the job failed")
-    config_in: Dict[str, Any] = Field(
-        default_factory=dict, description="Configuration information (e.g. 'description') to apply to model."
-    )
-    config_out: Optional[AnyModelConfig] = Field(
-        default=None, description="After successful installation, this will hold the configuration object."
-    )
-    inplace: bool = Field(
-        default=False, description="Leave model in its current location; otherwise install under models directory"
-    )
-    source: ModelSource = Field(description="Source (URL, repo_id, or local path) of model")
-    local_path: Path = Field(description="Path to locally-downloaded model; may be the same as the source")
-    bytes: int = Field(
-        default=0, description="For a remote model, the number of bytes downloaded so far (may not be available)"
-    )
-    total_bytes: int = Field(default=0, description="Total size of the model to be installed")
-    source_metadata: Optional[AnyModelRepoMetadata] = Field(
-        default=None, description="Metadata provided by the model source"
-    )
-    download_parts: Set[DownloadJob] = Field(
-        default_factory=set, description="Download jobs contributing to this install"
-    )
-    error: Optional[str] = Field(
-        default=None, description="On an error condition, this field will contain the text of the exception"
-    )
-    error_traceback: Optional[str] = Field(
-        default=None, description="On an error condition, this field will contain the exception traceback"
-    )
-    # internal flags and transitory settings
-    _install_tmpdir: Optional[Path] = PrivateAttr(default=None)
-    _exception: Optional[Exception] = PrivateAttr(default=None)
-
-    def set_error(self, e: Exception) -> None:
-        """Record the error and traceback from an exception."""
-        self._exception = e
-        self.error = str(e)
-        self.error_traceback = self._format_error(e)
-        self.status = InstallStatus.ERROR
-        self.error_reason = self._exception.__class__.__name__ if self._exception else None
-
-    def cancel(self) -> None:
-        """Call to cancel the job."""
-        self.status = InstallStatus.CANCELLED
-
-    @property
-    def error_type(self) -> Optional[str]:
-        """Class name of the exception that led to status==ERROR."""
-        return self._exception.__class__.__name__ if self._exception else None
-
-    def _format_error(self, exception: Exception) -> str:
-        """Error traceback."""
-        return "".join(traceback.format_exception(exception))
-
-    @property
-    def cancelled(self) -> bool:
-        """Set status to CANCELLED."""
-        return self.status == InstallStatus.CANCELLED
-
-    @property
-    def errored(self) -> bool:
-        """Return true if job has errored."""
-        return self.status == InstallStatus.ERROR
-
-    @property
-    def waiting(self) -> bool:
-        """Return true if job is waiting to run."""
-        return self.status == InstallStatus.WAITING
-
-    @property
-    def downloading(self) -> bool:
-        """Return true if job is downloading."""
-        return self.status == InstallStatus.DOWNLOADING
-
-    @property
-    def downloads_done(self) -> bool:
-        """Return true if job's downloads ae done."""
-        return self.status == InstallStatus.DOWNLOADS_DONE
-
-    @property
-    def running(self) -> bool:
-        """Return true if job is running."""
-        return self.status == InstallStatus.RUNNING
-
-    @property
-    def complete(self) -> bool:
-        """Return true if job completed without errors."""
-        return self.status == InstallStatus.COMPLETED
-
-    @property
-    def in_terminal_state(self) -> bool:
-        """Return true if job is in a terminal state."""
-        return self.status in [InstallStatus.COMPLETED, InstallStatus.ERROR, InstallStatus.CANCELLED]
+from invokeai.backend.model_manager import AnyModelConfig
 
 
 class ModelInstallServiceBase(ABC):
@@ -282,7 +57,7 @@ class ModelInstallServiceBase(ABC):
 
     @property
     @abstractmethod
-    def event_bus(self) -> Optional[EventServiceBase]:
+    def event_bus(self) -> Optional["EventServiceBase"]:
         """Return the event service base object associated with the installer."""
 
     @abstractmethod
@@ -468,12 +243,11 @@ class ModelInstallServiceBase(ABC):
         """
 
     @abstractmethod
-    def download_and_cache(self, source: Union[str, AnyHttpUrl], access_token: Optional[str] = None) -> Path:
+    def download_and_cache_model(self, source: str | AnyHttpUrl) -> Path:
         """
         Download the model file located at source to the models cache and return its Path.
 
-        :param source: A Url or a string that can be converted into one.
-        :param access_token: Optional access token to access restricted resources.
+        :param source: A string representing a URL or repo_id.
 
         The model file will be downloaded into the system-wide model cache
         (`models/.cache`) if it isn't already there. Note that the model cache

invokeai/app/services/model_install/model_install_common.py

@@ -0,0 +1,227 @@
+import re
+import traceback
+from enum import Enum
+from pathlib import Path
+from typing import Any, Dict, Literal, Optional, Set, Union
+
+from pydantic import BaseModel, Field, PrivateAttr, field_validator
+from pydantic.networks import AnyHttpUrl
+from typing_extensions import Annotated
+
+from invokeai.app.services.download import DownloadJob, MultiFileDownloadJob
+from invokeai.backend.model_manager import AnyModelConfig, ModelRepoVariant
+from invokeai.backend.model_manager.config import ModelSourceType
+from invokeai.backend.model_manager.metadata import AnyModelRepoMetadata
+
+
+class InstallStatus(str, Enum):
+    """State of an install job running in the background."""
+
+    WAITING = "waiting"  # waiting to be dequeued
+    DOWNLOADING = "downloading"  # downloading of model files in process
+    DOWNLOADS_DONE = "downloads_done"  # downloading done, waiting to run
+    RUNNING = "running"  # being processed
+    COMPLETED = "completed"  # finished running
+    ERROR = "error"  # terminated with an error message
+    CANCELLED = "cancelled"  # terminated with an error message
+
+
+class UnknownInstallJobException(Exception):
+    """Raised when the status of an unknown job is requested."""
+
+
+class StringLikeSource(BaseModel):
+    """
+    Base class for model sources, implements functions that lets the source be sorted and indexed.
+
+    These shenanigans let this stuff work:
+
+      source1 = LocalModelSource(path='C:/users/mort/foo.safetensors')
+      mydict = {source1: 'model 1'}
+      assert mydict['C:/users/mort/foo.safetensors'] == 'model 1'
+      assert mydict[LocalModelSource(path='C:/users/mort/foo.safetensors')] == 'model 1'
+
+      source2 = LocalModelSource(path=Path('C:/users/mort/foo.safetensors'))
+      assert source1 == source2
+      assert source1 == 'C:/users/mort/foo.safetensors'
+    """
+
+    def __hash__(self) -> int:
+        """Return hash of the path field, for indexing."""
+        return hash(str(self))
+
+    def __lt__(self, other: object) -> int:
+        """Return comparison of the stringified version, for sorting."""
+        return str(self) < str(other)
+
+    def __eq__(self, other: object) -> bool:
+        """Return equality on the stringified version."""
+        if isinstance(other, Path):
+            return str(self) == other.as_posix()
+        else:
+            return str(self) == str(other)
+
+
+class LocalModelSource(StringLikeSource):
+    """A local file or directory path."""
+
+    path: str | Path
+    inplace: Optional[bool] = False
+    type: Literal["local"] = "local"
+
+    # these methods allow the source to be used in a string-like way,
+    # for example as an index into a dict
+    def __str__(self) -> str:
+        """Return string version of path when string rep needed."""
+        return Path(self.path).as_posix()
+
+
+class HFModelSource(StringLikeSource):
+    """
+    A HuggingFace repo_id with optional variant, sub-folder and access token.
+    Note that the variant option, if not provided to the constructor, will default to fp16, which is
+    what people (almost) always want.
+    """
+
+    repo_id: str
+    variant: Optional[ModelRepoVariant] = ModelRepoVariant.FP16
+    subfolder: Optional[Path] = None
+    access_token: Optional[str] = None
+    type: Literal["hf"] = "hf"
+
+    @field_validator("repo_id")
+    @classmethod
+    def proper_repo_id(cls, v: str) -> str:  # noqa D102
+        if not re.match(r"^([.\w-]+/[.\w-]+)$", v):
+            raise ValueError(f"{v}: invalid repo_id format")
+        return v
+
+    def __str__(self) -> str:
+        """Return string version of repoid when string rep needed."""
+        base: str = self.repo_id
+        if self.variant:
+            base += f":{self.variant or ''}"
+        if self.subfolder:
+            base += f":{self.subfolder}"
+        return base
+
+
+class URLModelSource(StringLikeSource):
+    """A generic URL point to a checkpoint file."""
+
+    url: AnyHttpUrl
+    access_token: Optional[str] = None
+    type: Literal["url"] = "url"
+
+    def __str__(self) -> str:
+        """Return string version of the url when string rep needed."""
+        return str(self.url)
+
+
+ModelSource = Annotated[Union[LocalModelSource, HFModelSource, URLModelSource], Field(discriminator="type")]
+
+MODEL_SOURCE_TO_TYPE_MAP = {
+    URLModelSource: ModelSourceType.Url,
+    HFModelSource: ModelSourceType.HFRepoID,
+    LocalModelSource: ModelSourceType.Path,
+}
+
+
+class ModelInstallJob(BaseModel):
+    """Object that tracks the current status of an install request."""
+
+    id: int = Field(description="Unique ID for this job")
+    status: InstallStatus = Field(default=InstallStatus.WAITING, description="Current status of install process")
+    error_reason: Optional[str] = Field(default=None, description="Information about why the job failed")
+    config_in: Dict[str, Any] = Field(
+        default_factory=dict, description="Configuration information (e.g. 'description') to apply to model."
+    )
+    config_out: Optional[AnyModelConfig] = Field(
+        default=None, description="After successful installation, this will hold the configuration object."
+    )
+    inplace: bool = Field(
+        default=False, description="Leave model in its current location; otherwise install under models directory"
+    )
+    source: ModelSource = Field(description="Source (URL, repo_id, or local path) of model")
+    local_path: Path = Field(description="Path to locally-downloaded model; may be the same as the source")
+    bytes: int = Field(
+        default=0, description="For a remote model, the number of bytes downloaded so far (may not be available)"
+    )
+    total_bytes: int = Field(default=0, description="Total size of the model to be installed")
+    source_metadata: Optional[AnyModelRepoMetadata] = Field(
+        default=None, description="Metadata provided by the model source"
+    )
+    download_parts: Set[DownloadJob] = Field(
+        default_factory=set, description="Download jobs contributing to this install"
+    )
+    error: Optional[str] = Field(
+        default=None, description="On an error condition, this field will contain the text of the exception"
+    )
+    error_traceback: Optional[str] = Field(
+        default=None, description="On an error condition, this field will contain the exception traceback"
+    )
+    # internal flags and transitory settings
+    _install_tmpdir: Optional[Path] = PrivateAttr(default=None)
+    _multifile_job: Optional[MultiFileDownloadJob] = PrivateAttr(default=None)
+    _exception: Optional[Exception] = PrivateAttr(default=None)
+
+    def set_error(self, e: Exception) -> None:
+        """Record the error and traceback from an exception."""
+        self._exception = e
+        self.error = str(e)
+        self.error_traceback = self._format_error(e)
+        self.status = InstallStatus.ERROR
+        self.error_reason = self._exception.__class__.__name__ if self._exception else None
+
+    def cancel(self) -> None:
+        """Call to cancel the job."""
+        self.status = InstallStatus.CANCELLED
+
+    @property
+    def error_type(self) -> Optional[str]:
+        """Class name of the exception that led to status==ERROR."""
+        return self._exception.__class__.__name__ if self._exception else None
+
+    def _format_error(self, exception: Exception) -> str:
+        """Error traceback."""
+        return "".join(traceback.format_exception(exception))
+
+    @property
+    def cancelled(self) -> bool:
+        """Set status to CANCELLED."""
+        return self.status == InstallStatus.CANCELLED
+
+    @property
+    def errored(self) -> bool:
+        """Return true if job has errored."""
+        return self.status == InstallStatus.ERROR
+
+    @property
+    def waiting(self) -> bool:
+        """Return true if job is waiting to run."""
+        return self.status == InstallStatus.WAITING
+
+    @property
+    def downloading(self) -> bool:
+        """Return true if job is downloading."""
+        return self.status == InstallStatus.DOWNLOADING
+
+    @property
+    def downloads_done(self) -> bool:
+        """Return true if job's downloads ae done."""
+        return self.status == InstallStatus.DOWNLOADS_DONE
+
+    @property
+    def running(self) -> bool:
+        """Return true if job is running."""
+        return self.status == InstallStatus.RUNNING
+
+    @property
+    def complete(self) -> bool:
+        """Return true if job completed without errors."""
+        return self.status == InstallStatus.COMPLETED
+
+    @property
+    def in_terminal_state(self) -> bool:
+        """Return true if job is in a terminal state."""
+        return self.status in [InstallStatus.COMPLETED, InstallStatus.ERROR, InstallStatus.CANCELLED]

invokeai/app/services/model_install/model_install_default.py

@@ -5,23 +5,24 @@ import os
 import re
 import threading
 import time
-from hashlib import sha256
 from pathlib import Path
 from queue import Empty, Queue
 from shutil import copyfile, copytree, move, rmtree
 from tempfile import mkdtemp
-from typing import Any, Dict, List, Optional, Union
+from typing import Any, Dict, List, Optional, Tuple, Type, Union
 
 import torch
 import yaml
 from huggingface_hub import HfFolder
 from pydantic.networks import AnyHttpUrl
+from pydantic_core import Url
 from requests import Session
 
 from invokeai.app.services.config import InvokeAIAppConfig
-from invokeai.app.services.download import DownloadJob, DownloadQueueServiceBase, TqdmProgress
+from invokeai.app.services.download import DownloadQueueServiceBase, MultiFileDownloadJob
 from invokeai.app.services.events.events_base import EventServiceBase
 from invokeai.app.services.invoker import Invoker
+from invokeai.app.services.model_install.model_install_base import ModelInstallServiceBase
 from invokeai.app.services.model_records import DuplicateModelException, ModelRecordServiceBase
 from invokeai.app.services.model_records.model_records_base import ModelRecordChanges
 from invokeai.backend.model_manager.config import (
@@ -44,14 +45,14 @@ from invokeai.backend.model_manager.search import ModelSearch
 from invokeai.backend.util import InvokeAILogger
 from invokeai.backend.util.catch_sigint import catch_sigint
 from invokeai.backend.util.devices import TorchDevice
+from invokeai.backend.util.util import slugify
 
-from .model_install_base import (
+from .model_install_common import (
     MODEL_SOURCE_TO_TYPE_MAP,
     HFModelSource,
     InstallStatus,
     LocalModelSource,
     ModelInstallJob,
-    ModelInstallServiceBase,
     ModelSource,
     StringLikeSource,
     URLModelSource,
@@ -68,7 +69,7 @@ class ModelInstallService(ModelInstallServiceBase):
         app_config: InvokeAIAppConfig,
         record_store: ModelRecordServiceBase,
         download_queue: DownloadQueueServiceBase,
-        event_bus: Optional[EventServiceBase] = None,
+        event_bus: Optional["EventServiceBase"] = None,
         session: Optional[Session] = None,
     ):
         """
@@ -89,7 +90,7 @@ class ModelInstallService(ModelInstallServiceBase):
         self._downloads_changed_event = threading.Event()
         self._install_completed_event = threading.Event()
         self._download_queue = download_queue
-        self._download_cache: Dict[AnyHttpUrl, ModelInstallJob] = {}
+        self._download_cache: Dict[int, ModelInstallJob] = {}
         self._running = False
         self._session = session
         self._install_thread: Optional[threading.Thread] = None
@@ -104,7 +105,7 @@ class ModelInstallService(ModelInstallServiceBase):
         return self._record_store
 
     @property
-    def event_bus(self) -> Optional[EventServiceBase]:  # noqa D102
+    def event_bus(self) -> Optional["EventServiceBase"]:  # noqa D102
         return self._event_bus
 
     # make the invoker optional here because we don't need it and it
@@ -208,33 +209,12 @@ class ModelInstallService(ModelInstallServiceBase):
         access_token: Optional[str] = None,
         inplace: Optional[bool] = False,
     ) -> ModelInstallJob:
-        variants = "|".join(ModelRepoVariant.__members__.values())
-        hf_repoid_re = f"^([^/:]+/[^/:]+)(?::({variants})?(?::/?([^:]+))?)?$"
-        source_obj: Optional[StringLikeSource] = None
-
-        if Path(source).exists():  # A local file or directory
-            source_obj = LocalModelSource(path=Path(source), inplace=inplace)
-        elif match := re.match(hf_repoid_re, source):
-            source_obj = HFModelSource(
-                repo_id=match.group(1),
-                variant=match.group(2) if match.group(2) else None,  # pass None rather than ''
-                subfolder=Path(match.group(3)) if match.group(3) else None,
-                access_token=access_token,
-            )
-        elif re.match(r"^https?://[^/]+", source):
-            # Pull the token from config if it exists and matches the URL
-            _token = access_token
-            if _token is None:
-                for pair in self.app_config.remote_api_tokens or []:
-                    if re.search(pair.url_regex, source):
-                        _token = pair.token
-                        break
-            source_obj = URLModelSource(
-                url=AnyHttpUrl(source),
-                access_token=_token,
-            )
-        else:
-            raise ValueError(f"Unsupported model source: '{source}'")
+        """Install a model using pattern matching to infer the type of source."""
+        source_obj = self._guess_source(source)
+        if isinstance(source_obj, LocalModelSource):
+            source_obj.inplace = inplace
+        elif isinstance(source_obj, HFModelSource) or isinstance(source_obj, URLModelSource):
+            source_obj.access_token = access_token
         return self.import_model(source_obj, config)
 
     def import_model(self, source: ModelSource, config: Optional[Dict[str, Any]] = None) -> ModelInstallJob:  # noqa D102
@@ -295,17 +275,23 @@ class ModelInstallService(ModelInstallServiceBase):
     def cancel_job(self, job: ModelInstallJob) -> None:
         """Cancel the indicated job."""
         job.cancel()
-        with self._lock:
-            self._cancel_download_parts(job)
+        self._logger.warning(f"Cancelling {job.source}")
+        if dj := job._multifile_job:
+            self._download_queue.cancel_job(dj)
 
     def prune_jobs(self) -> None:
         """Prune all completed and errored jobs."""
         unfinished_jobs = [x for x in self._install_jobs if not x.in_terminal_state]
         self._install_jobs = unfinished_jobs
 
-    def _migrate_yaml(self) -> None:
+    def _migrate_yaml(self, rename_yaml: Optional[bool] = True, overwrite_db: Optional[bool] = False) -> None:
         db_models = self.record_store.all_models()
 
+        if overwrite_db:
+            for model in db_models:
+                self.record_store.del_model(model.key)
+            db_models = self.record_store.all_models()
+
         legacy_models_yaml_path = (
             self._app_config.legacy_models_yaml_path or self._app_config.root_path / "configs" / "models.yaml"
         )
@@ -344,7 +330,7 @@ class ModelInstallService(ModelInstallServiceBase):
                     legacy_config_path = stanza.get("config")
                     if legacy_config_path:
                         # In v3, these paths were relative to the root. Migrate them to be relative to the legacy_conf_dir.
-                        legacy_config_path: Path = self._app_config.root_path / legacy_config_path
+                        legacy_config_path = self._app_config.root_path / legacy_config_path
                         if legacy_config_path.is_relative_to(self._app_config.legacy_conf_path):
                             legacy_config_path = legacy_config_path.relative_to(self._app_config.legacy_conf_path)
                         config["config_path"] = str(legacy_config_path)
@@ -355,7 +341,8 @@ class ModelInstallService(ModelInstallServiceBase):
                         self._logger.warning(f"Model at {model_path} could not be migrated: {e}")
 
             # Rename `models.yaml` to `models.yaml.bak` to prevent re-migration
-            legacy_models_yaml_path.rename(legacy_models_yaml_path.with_suffix(".yaml.bak"))
+            if rename_yaml:
+                legacy_models_yaml_path.rename(legacy_models_yaml_path.with_suffix(".yaml.bak"))
 
         # Unset the path - we are done with it either way
         self._app_config.legacy_models_yaml_path = None
@@ -384,38 +371,95 @@ class ModelInstallService(ModelInstallServiceBase):
             rmtree(model_path)
         self.unregister(key)
 
-    def download_and_cache(
+    @classmethod
+    def _download_cache_path(cls, source: Union[str, AnyHttpUrl], app_config: InvokeAIAppConfig) -> Path:
+        escaped_source = slugify(str(source))
+        return app_config.download_cache_path / escaped_source
+
+    def download_and_cache_model(
         self,
-        source: Union[str, AnyHttpUrl],
-        access_token: Optional[str] = None,
-        timeout: int = 0,
+        source: str | AnyHttpUrl,
     ) -> Path:
         """Download the model file located at source to the models cache and return its Path."""
-        model_hash = sha256(str(source).encode("utf-8")).hexdigest()[0:32]
-        model_path = self._app_config.convert_cache_path / model_hash
+        model_path = self._download_cache_path(str(source), self._app_config)
 
-        # We expect the cache directory to contain one and only one downloaded file.
+        # We expect the cache directory to contain one and only one downloaded file or directory.
         # We don't know the file's name in advance, as it is set by the download
         # content-disposition header.
         if model_path.exists():
-            contents = [x for x in model_path.iterdir() if x.is_file()]
+            contents: List[Path] = list(model_path.iterdir())
             if len(contents) > 0:
                 return contents[0]
 
         model_path.mkdir(parents=True, exist_ok=True)
-        job = self._download_queue.download(
-            source=AnyHttpUrl(str(source)),
+        model_source = self._guess_source(str(source))
+        remote_files, _ = self._remote_files_from_source(model_source)
+        job = self._multifile_download(
             dest=model_path,
-            access_token=access_token,
-            on_progress=TqdmProgress().update,
+            remote_files=remote_files,
+            subfolder=model_source.subfolder if isinstance(model_source, HFModelSource) else None,
         )
-        self._download_queue.wait_for_job(job, timeout)
+        files_string = "file" if len(remote_files) == 1 else "files"
+        self._logger.info(f"Queuing model download: {source} ({len(remote_files)} {files_string})")
+        self._download_queue.wait_for_job(job)
         if job.complete:
             assert job.download_path is not None
             return job.download_path
         else:
             raise Exception(job.error)
 
+    def _remote_files_from_source(
+        self, source: ModelSource
+    ) -> Tuple[List[RemoteModelFile], Optional[AnyModelRepoMetadata]]:
+        metadata = None
+        if isinstance(source, HFModelSource):
+            metadata = HuggingFaceMetadataFetch(self._session).from_id(source.repo_id, source.variant)
+            assert isinstance(metadata, ModelMetadataWithFiles)
+            return (
+                metadata.download_urls(
+                    variant=source.variant or self._guess_variant(),
+                    subfolder=source.subfolder,
+                    session=self._session,
+                ),
+                metadata,
+            )
+
+        if isinstance(source, URLModelSource):
+            try:
+                fetcher = self.get_fetcher_from_url(str(source.url))
+                kwargs: dict[str, Any] = {"session": self._session}
+                metadata = fetcher(**kwargs).from_url(source.url)
+                assert isinstance(metadata, ModelMetadataWithFiles)
+                return metadata.download_urls(session=self._session), metadata
+            except ValueError:
+                pass
+
+            return [RemoteModelFile(url=source.url, path=Path("."), size=0)], None
+
+        raise Exception(f"No files associated with {source}")
+
+    def _guess_source(self, source: str) -> ModelSource:
+        """Turn a source string into a ModelSource object."""
+        variants = "|".join(ModelRepoVariant.__members__.values())
+        hf_repoid_re = f"^([^/:]+/[^/:]+)(?::({variants})?(?::/?([^:]+))?)?$"
+        source_obj: Optional[StringLikeSource] = None
+
+        if Path(source).exists():  # A local file or directory
+            source_obj = LocalModelSource(path=Path(source))
+        elif match := re.match(hf_repoid_re, source):
+            source_obj = HFModelSource(
+                repo_id=match.group(1),
+                variant=ModelRepoVariant(match.group(2)) if match.group(2) else None,  # pass None rather than ''
+                subfolder=Path(match.group(3)) if match.group(3) else None,
+            )
+        elif re.match(r"^https?://[^/]+", source):
+            source_obj = URLModelSource(
+                url=Url(source),
+            )
+        else:
+            raise ValueError(f"Unsupported model source: '{source}'")
+        return source_obj
+
     # --------------------------------------------------------------------------------------------
     # Internal functions that manage the installer threads
     # --------------------------------------------------------------------------------------------
@@ -476,16 +520,19 @@ class ModelInstallService(ModelInstallServiceBase):
         job.config_out = self.record_store.get_model(key)
         self._signal_job_completed(job)
 
-    def _set_error(self, job: ModelInstallJob, excp: Exception) -> None:
-        if any(x.content_type is not None and "text/html" in x.content_type for x in job.download_parts):
-            job.set_error(
+    def _set_error(self, install_job: ModelInstallJob, excp: Exception) -> None:
+        multifile_download_job = install_job._multifile_job
+        if multifile_download_job and any(
+            x.content_type is not None and "text/html" in x.content_type for x in multifile_download_job.download_parts
+        ):
+            install_job.set_error(
                 InvalidModelConfigException(
-                    f"At least one file in {job.local_path} is an HTML page, not a model. This can happen when an access token is required to download."
+                    f"At least one file in {install_job.local_path} is an HTML page, not a model. This can happen when an access token is required to download."
                 )
             )
         else:
-            job.set_error(excp)
-        self._signal_job_errored(job)
+            install_job.set_error(excp)
+        self._signal_job_errored(install_job)
 
     # --------------------------------------------------------------------------------------------
     # Internal functions that manage the models directory
@@ -511,7 +558,6 @@ class ModelInstallService(ModelInstallServiceBase):
         This is typically only used during testing with a new DB or when using the memory DB, because those are the
         only situations in which we may have orphaned models in the models directory.
         """
-
         installed_model_paths = {
             (self._app_config.models_path / x.path).resolve() for x in self.record_store.all_models()
         }
@@ -523,8 +569,13 @@ class ModelInstallService(ModelInstallServiceBase):
             if resolved_path in installed_model_paths:
                 return True
             # Skip core models entirely - these aren't registered with the model manager.
-            if str(resolved_path).startswith(str(self.app_config.models_path / "core")):
-                return False
+            for special_directory in [
+                self.app_config.models_path / "core",
+                self.app_config.convert_cache_dir,
+                self.app_config.download_cache_dir,
+            ]:
+                if resolved_path.is_relative_to(special_directory):
+                    return False
             try:
                 model_id = self.register_path(model_path)
                 self._logger.info(f"Registered {model_path.name} with id {model_id}")
@@ -639,20 +690,15 @@ class ModelInstallService(ModelInstallServiceBase):
             inplace=source.inplace or False,
         )
 
-    def _import_from_hf(self, source: HFModelSource, config: Optional[Dict[str, Any]]) -> ModelInstallJob:
+    def _import_from_hf(
+        self,
+        source: HFModelSource,
+        config: Optional[Dict[str, Any]] = None,
+    ) -> ModelInstallJob:
         # Add user's cached access token to HuggingFace requests
-        source.access_token = source.access_token or HfFolder.get_token()
-        if not source.access_token:
-            self._logger.info("No HuggingFace access token present; some models may not be downloadable.")
-
-        metadata = HuggingFaceMetadataFetch(self._session).from_id(source.repo_id, source.variant)
-        assert isinstance(metadata, ModelMetadataWithFiles)
-        remote_files = metadata.download_urls(
-            variant=source.variant or self._guess_variant(),
-            subfolder=source.subfolder,
-            session=self._session,
-        )
-
+        if source.access_token is None:
+            source.access_token = HfFolder.get_token()
+        remote_files, metadata = self._remote_files_from_source(source)
         return self._import_remote_model(
             source=source,
             config=config,
@@ -660,22 +706,12 @@ class ModelInstallService(ModelInstallServiceBase):
             metadata=metadata,
         )
 
-    def _import_from_url(self, source: URLModelSource, config: Optional[Dict[str, Any]]) -> ModelInstallJob:
-        # URLs from HuggingFace will be handled specially
-        metadata = None
-        fetcher = None
-        try:
-            fetcher = self.get_fetcher_from_url(str(source.url))
-        except ValueError:
-            pass
-        kwargs: dict[str, Any] = {"session": self._session}
-        if fetcher is not None:
-            metadata = fetcher(**kwargs).from_url(source.url)
-        self._logger.debug(f"metadata={metadata}")
-        if metadata and isinstance(metadata, ModelMetadataWithFiles):
-            remote_files = metadata.download_urls(session=self._session)
-        else:
-            remote_files = [RemoteModelFile(url=source.url, path=Path("."), size=0)]
+    def _import_from_url(
+        self,
+        source: URLModelSource,
+        config: Optional[Dict[str, Any]],
+    ) -> ModelInstallJob:
+        remote_files, metadata = self._remote_files_from_source(source)
         return self._import_remote_model(
             source=source,
             config=config,
@@ -690,12 +726,9 @@ class ModelInstallService(ModelInstallServiceBase):
         metadata: Optional[AnyModelRepoMetadata],
         config: Optional[Dict[str, Any]],
     ) -> ModelInstallJob:
-        # TODO: Replace with tempfile.tmpdir() when multithreading is cleaned up.
-        # Currently the tmpdir isn't automatically removed at exit because it is
-        # being held in a daemon thread.
         if len(remote_files) == 0:
             raise ValueError(f"{source}: No downloadable files found")
-        tmpdir = Path(
+        destdir = Path(
             mkdtemp(
                 dir=self._app_config.models_path,
                 prefix=TMPDIR_PREFIX,
@@ -706,55 +739,28 @@ class ModelInstallService(ModelInstallServiceBase):
             source=source,
             config_in=config or {},
             source_metadata=metadata,
-            local_path=tmpdir,  # local path may change once the download has started due to content-disposition handling
+            local_path=destdir,  # local path may change once the download has started due to content-disposition handling
             bytes=0,
             total_bytes=0,
         )
-        # In the event that there is a subfolder specified in the source,
-        # we need to remove it from the destination path in order to avoid
-        # creating unwanted subfolders
-        if isinstance(source, HFModelSource) and source.subfolder:
-            root = Path(remote_files[0].path.parts[0])
-            subfolder = root / source.subfolder
-        else:
-            root = Path(".")
-            subfolder = Path(".")
+        # remember the temporary directory for later removal
+        install_job._install_tmpdir = destdir
+        install_job.total_bytes = sum((x.size or 0) for x in remote_files)
 
-        # we remember the path up to the top of the tmpdir so that it may be
-        # removed safely at the end of the install process.
-        install_job._install_tmpdir = tmpdir
-        assert install_job.total_bytes is not None  # to avoid type checking complaints in the loop below
+        multifile_job = self._multifile_download(
+            remote_files=remote_files,
+            dest=destdir,
+            subfolder=source.subfolder if isinstance(source, HFModelSource) else None,
+            access_token=source.access_token,
+            submit_job=False,  # Important! Don't submit the job until we have set our _download_cache dict
+        )
+        self._download_cache[multifile_job.id] = install_job
+        install_job._multifile_job = multifile_job
 
-        files_string = "file" if len(remote_files) == 1 else "file"
-        self._logger.info(f"Queuing model install: {source} ({len(remote_files)} {files_string})")
+        files_string = "file" if len(remote_files) == 1 else "files"
+        self._logger.info(f"Queueing model install: {source} ({len(remote_files)} {files_string})")
         self._logger.debug(f"remote_files={remote_files}")
-        for model_file in remote_files:
-            url = model_file.url
-            path = root / model_file.path.relative_to(subfolder)
-            self._logger.debug(f"Downloading {url} => {path}")
-            install_job.total_bytes += model_file.size
-            assert hasattr(source, "access_token")
-            dest = tmpdir / path.parent
-            dest.mkdir(parents=True, exist_ok=True)
-            download_job = DownloadJob(
-                source=url,
-                dest=dest,
-                access_token=source.access_token,
-            )
-            self._download_cache[download_job.source] = install_job  # matches a download job to an install job
-            install_job.download_parts.add(download_job)
-
-        # only start the jobs once install_job.download_parts is fully populated
-        for download_job in install_job.download_parts:
-            self._download_queue.submit_download_job(
-                download_job,
-                on_start=self._download_started_callback,
-                on_progress=self._download_progress_callback,
-                on_complete=self._download_complete_callback,
-                on_error=self._download_error_callback,
-                on_cancelled=self._download_cancelled_callback,
-            )
-
+        self._download_queue.submit_multifile_download(multifile_job)
         return install_job
 
     def _stat_size(self, path: Path) -> int:
@@ -766,87 +772,104 @@ class ModelInstallService(ModelInstallServiceBase):
                 size += sum(self._stat_size(Path(root, x)) for x in files)
         return size
 
+    def _multifile_download(
+        self,
+        remote_files: List[RemoteModelFile],
+        dest: Path,
+        subfolder: Optional[Path] = None,
+        access_token: Optional[str] = None,
+        submit_job: bool = True,
+    ) -> MultiFileDownloadJob:
+        # HuggingFace repo subfolders are a little tricky. If the name of the model is "sdxl-turbo", and
+        # we are installing the "vae" subfolder, we do not want to create an additional folder level, such
+        # as "sdxl-turbo/vae", nor do we want to put the contents of the vae folder directly into "sdxl-turbo".
+        # So what we do is to synthesize a folder named "sdxl-turbo_vae" here.
+        if subfolder:
+            top = Path(remote_files[0].path.parts[0])  # e.g. "sdxl-turbo/"
+            path_to_remove = top / subfolder.parts[-1]  # sdxl-turbo/vae/
+            path_to_add = Path(f"{top}_{subfolder}")
+        else:
+            path_to_remove = Path(".")
+            path_to_add = Path(".")
+
+        parts: List[RemoteModelFile] = []
+        for model_file in remote_files:
+            assert model_file.size is not None
+            parts.append(
+                RemoteModelFile(
+                    url=model_file.url,  # if a subfolder, then sdxl-turbo_vae/config.json
+                    path=path_to_add / model_file.path.relative_to(path_to_remove),
+                )
+            )
+
+        return self._download_queue.multifile_download(
+            parts=parts,
+            dest=dest,
+            access_token=access_token,
+            submit_job=submit_job,
+            on_start=self._download_started_callback,
+            on_progress=self._download_progress_callback,
+            on_complete=self._download_complete_callback,
+            on_error=self._download_error_callback,
+            on_cancelled=self._download_cancelled_callback,
+        )
+
     # ------------------------------------------------------------------
     # Callbacks are executed by the download queue in a separate thread
     # ------------------------------------------------------------------
-    def _download_started_callback(self, download_job: DownloadJob) -> None:
-        self._logger.info(f"Model download started: {download_job.source}")
+    def _download_started_callback(self, download_job: MultiFileDownloadJob) -> None:
         with self._lock:
-            install_job = self._download_cache[download_job.source]
-            install_job.status = InstallStatus.DOWNLOADING
+            if install_job := self._download_cache.get(download_job.id, None):
+                install_job.status = InstallStatus.DOWNLOADING
 
-            assert download_job.download_path
-            if install_job.local_path == install_job._install_tmpdir:
-                partial_path = download_job.download_path.relative_to(install_job._install_tmpdir)
-                dest_name = partial_path.parts[0]
-                install_job.local_path = install_job._install_tmpdir / dest_name
+                if install_job.local_path == install_job._install_tmpdir:  # first time
+                    assert download_job.download_path
+                    install_job.local_path = download_job.download_path
+                install_job.download_parts = download_job.download_parts
+                install_job.bytes = sum(x.bytes for x in download_job.download_parts)
+                install_job.total_bytes = download_job.total_bytes
+                self._signal_job_download_started(install_job)
 
-            # Update the total bytes count for remote sources.
-            if not install_job.total_bytes:
-                install_job.total_bytes = sum(x.total_bytes for x in install_job.download_parts)
-
-    def _download_progress_callback(self, download_job: DownloadJob) -> None:
+    def _download_progress_callback(self, download_job: MultiFileDownloadJob) -> None:
         with self._lock:
-            install_job = self._download_cache[download_job.source]
-            if install_job.cancelled:  # This catches the case in which the caller directly calls job.cancel()
-                self._cancel_download_parts(install_job)
-            else:
-                # update sizes
-                install_job.bytes = sum(x.bytes for x in install_job.download_parts)
-                self._signal_job_downloading(install_job)
+            if install_job := self._download_cache.get(download_job.id, None):
+                if install_job.cancelled:  # This catches the case in which the caller directly calls job.cancel()
+                    self._download_queue.cancel_job(download_job)
+                else:
+                    # update sizes
+                    install_job.bytes = sum(x.bytes for x in download_job.download_parts)
+                    install_job.total_bytes = sum(x.total_bytes for x in download_job.download_parts)
+                    self._signal_job_downloading(install_job)
 
-    def _download_complete_callback(self, download_job: DownloadJob) -> None:
-        self._logger.info(f"Model download complete: {download_job.source}")
+    def _download_complete_callback(self, download_job: MultiFileDownloadJob) -> None:
         with self._lock:
-            install_job = self._download_cache[download_job.source]
-
-            # are there any more active jobs left in this task?
-            if install_job.downloading and all(x.complete for x in install_job.download_parts):
+            if install_job := self._download_cache.pop(download_job.id, None):
                 self._signal_job_downloads_done(install_job)
-                self._put_in_queue(install_job)
+                self._put_in_queue(install_job)  # this starts the installation and registration
 
-            # Let other threads know that the number of downloads has changed
-            self._download_cache.pop(download_job.source, None)
-            self._downloads_changed_event.set()
+                # Let other threads know that the number of downloads has changed
+                self._downloads_changed_event.set()
 
-    def _download_error_callback(self, download_job: DownloadJob, excp: Optional[Exception] = None) -> None:
+    def _download_error_callback(self, download_job: MultiFileDownloadJob, excp: Optional[Exception] = None) -> None:
         with self._lock:
-            install_job = self._download_cache.pop(download_job.source, None)
-            assert install_job is not None
-            assert excp is not None
-            install_job.set_error(excp)
-            self._logger.error(
-                f"Cancelling {install_job.source} due to an error while downloading {download_job.source}: {str(excp)}"
-            )
-            self._cancel_download_parts(install_job)
+            if install_job := self._download_cache.pop(download_job.id, None):
+                assert excp is not None
+                install_job.set_error(excp)
+                self._download_queue.cancel_job(download_job)
 
-            # Let other threads know that the number of downloads has changed
-            self._downloads_changed_event.set()
+                # Let other threads know that the number of downloads has changed
+                self._downloads_changed_event.set()
 
-    def _download_cancelled_callback(self, download_job: DownloadJob) -> None:
+    def _download_cancelled_callback(self, download_job: MultiFileDownloadJob) -> None:
         with self._lock:
-            install_job = self._download_cache.pop(download_job.source, None)
-            if not install_job:
-                return
-            self._downloads_changed_event.set()
-            self._logger.warning(f"Model download canceled: {download_job.source}")
-            # if install job has already registered an error, then do not replace its status with cancelled
-            if not install_job.errored:
-                install_job.cancel()
-            self._cancel_download_parts(install_job)
+            if install_job := self._download_cache.pop(download_job.id, None):
+                self._downloads_changed_event.set()
+                # if install job has already registered an error, then do not replace its status with cancelled
+                if not install_job.errored:
+                    install_job.cancel()
 
-            # Let other threads know that the number of downloads has changed
-            self._downloads_changed_event.set()
-
-    def _cancel_download_parts(self, install_job: ModelInstallJob) -> None:
-        # on multipart downloads, _cancel_components() will get called repeatedly from the download callbacks
-        # do not lock here because it gets called within a locked context
-        for s in install_job.download_parts:
-            self._download_queue.cancel_job(s)
-
-        if all(x.in_terminal_state for x in install_job.download_parts):
-            # When all parts have reached their terminal state, we finalize the job to clean up the temporary directory and other resources
-            self._put_in_queue(install_job)
+                # Let other threads know that the number of downloads has changed
+                self._downloads_changed_event.set()
 
     # ------------------------------------------------------------------------------------------------
     # Internal methods that put events on the event bus
@@ -855,35 +878,27 @@ class ModelInstallService(ModelInstallServiceBase):
         job.status = InstallStatus.RUNNING
         self._logger.info(f"Model install started: {job.source}")
         if self._event_bus:
-            self._event_bus.emit_model_install_running(str(job.source))
+            self._event_bus.emit_model_install_started(job)
+
+    def _signal_job_download_started(self, job: ModelInstallJob) -> None:
+        if self._event_bus:
+            assert job._multifile_job is not None
+            assert job.bytes is not None
+            assert job.total_bytes is not None
+            self._event_bus.emit_model_install_download_started(job)
 
     def _signal_job_downloading(self, job: ModelInstallJob) -> None:
         if self._event_bus:
-            parts: List[Dict[str, str | int]] = [
-                {
-                    "url": str(x.source),
-                    "local_path": str(x.download_path),
-                    "bytes": x.bytes,
-                    "total_bytes": x.total_bytes,
-                }
-                for x in job.download_parts
-            ]
+            assert job._multifile_job is not None
             assert job.bytes is not None
             assert job.total_bytes is not None
-            self._event_bus.emit_model_install_downloading(
-                str(job.source),
-                local_path=job.local_path.as_posix(),
-                parts=parts,
-                bytes=job.bytes,
-                total_bytes=job.total_bytes,
-                id=job.id,
-            )
+            self._event_bus.emit_model_install_download_progress(job)
 
     def _signal_job_downloads_done(self, job: ModelInstallJob) -> None:
         job.status = InstallStatus.DOWNLOADS_DONE
         self._logger.info(f"Model download complete: {job.source}")
         if self._event_bus:
-            self._event_bus.emit_model_install_downloads_done(str(job.source))
+            self._event_bus.emit_model_install_downloads_complete(job)
 
     def _signal_job_completed(self, job: ModelInstallJob) -> None:
         job.status = InstallStatus.COMPLETED
@@ -893,25 +908,28 @@ class ModelInstallService(ModelInstallServiceBase):
         if self._event_bus:
             assert job.local_path is not None
             assert job.config_out is not None
-            key = job.config_out.key
-            self._event_bus.emit_model_install_completed(str(job.source), key, id=job.id)
+            self._event_bus.emit_model_install_complete(job)
 
     def _signal_job_errored(self, job: ModelInstallJob) -> None:
         self._logger.error(f"Model install error: {job.source}\n{job.error_type}: {job.error}")
         if self._event_bus:
-            error_type = job.error_type
-            error = job.error
-            assert error_type is not None
-            assert error is not None
-            self._event_bus.emit_model_install_error(str(job.source), error_type, error, id=job.id)
+            assert job.error_type is not None
+            assert job.error is not None
+            self._event_bus.emit_model_install_error(job)
 
     def _signal_job_cancelled(self, job: ModelInstallJob) -> None:
         self._logger.info(f"Model install canceled: {job.source}")
         if self._event_bus:
-            self._event_bus.emit_model_install_cancelled(str(job.source), id=job.id)
+            self._event_bus.emit_model_install_cancelled(job)
 
     @staticmethod
-    def get_fetcher_from_url(url: str) -> ModelMetadataFetchBase:
+    def get_fetcher_from_url(url: str) -> Type[ModelMetadataFetchBase]:
+        """
+        Return a metadata fetcher appropriate for provided url.
+
+        This used to be more useful, but the number of supported model
+        sources has been reduced to HuggingFace alone.
+        """
         if re.match(r"^https?://huggingface.co/[^/]+/[^/]+$", url.lower()):
             return HuggingFaceMetadataFetch
         raise ValueError(f"Unsupported model source: '{url}'")

invokeai/app/services/model_load/model_load_base.py

@@ -2,11 +2,11 @@
 """Base class for model loader."""
 
 from abc import ABC, abstractmethod
-from typing import Optional
+from pathlib import Path
+from typing import Callable, Optional
 
-from invokeai.app.services.shared.invocation_context import InvocationContextData
 from invokeai.backend.model_manager import AnyModel, AnyModelConfig, SubModelType
-from invokeai.backend.model_manager.load import LoadedModel
+from invokeai.backend.model_manager.load import LoadedModel, LoadedModelWithoutConfig
 from invokeai.backend.model_manager.load.convert_cache import ModelConvertCacheBase
 from invokeai.backend.model_manager.load.model_cache.model_cache_base import ModelCacheBase
 
@@ -15,18 +15,12 @@ class ModelLoadServiceBase(ABC):
     """Wrapper around AnyModelLoader."""
 
     @abstractmethod
-    def load_model(
-        self,
-        model_config: AnyModelConfig,
-        submodel_type: Optional[SubModelType] = None,
-        context_data: Optional[InvocationContextData] = None,
-    ) -> LoadedModel:
+    def load_model(self, model_config: AnyModelConfig, submodel_type: Optional[SubModelType] = None) -> LoadedModel:
         """
         Given a model's configuration, load it and return the LoadedModel object.
 
         :param model_config: Model configuration record (as returned by ModelRecordBase.get_model())
         :param submodel: For main (pipeline models), the submodel to fetch.
-        :param context_data: Invocation context data used for event reporting
         """
 
     @property
@@ -38,3 +32,31 @@ class ModelLoadServiceBase(ABC):
     @abstractmethod
     def convert_cache(self) -> ModelConvertCacheBase:
         """Return the checkpoint convert cache used by this loader."""
+
+    @property
+    @abstractmethod
+    def gpu_count(self) -> int:
+        """Return the number of GPUs we are configured to use."""
+
+    @abstractmethod
+    def load_model_from_path(
+        self, model_path: Path, loader: Optional[Callable[[Path], AnyModel]] = None
+    ) -> LoadedModelWithoutConfig:
+        """
+        Load the model file or directory located at the indicated Path.
+
+        This will load an arbitrary model file into the RAM cache. If the optional loader
+        argument is provided, the loader will be invoked to load the model into
+        memory. Otherwise the method will call safetensors.torch.load_file() or
+        torch.load() as appropriate to the file suffix.
+
+        Be aware that this returns a LoadedModelWithoutConfig object, which is the same as
+        LoadedModel, but without the config attribute.
+
+        Args:
+          model_path: A pathlib.Path to a checkpoint-style models file
+          loader: A Callable that expects a Path and returns a Dict[str, Tensor]
+
+        Returns:
+          A LoadedModel object.
+        """

invokeai/app/services/model_load/model_load_default.py

@@ -1,19 +1,26 @@
 # Copyright (c) 2024 Lincoln D. Stein and the InvokeAI Team
 """Implementation of model loader service."""
 
-from typing import Optional, Type
+from pathlib import Path
+from typing import Callable, Optional, Type
+
+from picklescan.scanner import scan_file_path
+from safetensors.torch import load_file as safetensors_load_file
+from torch import load as torch_load
 
 from invokeai.app.services.config import InvokeAIAppConfig
 from invokeai.app.services.invoker import Invoker
-from invokeai.app.services.shared.invocation_context import InvocationContextData
 from invokeai.backend.model_manager import AnyModel, AnyModelConfig, SubModelType
 from invokeai.backend.model_manager.load import (
     LoadedModel,
+    LoadedModelWithoutConfig,
     ModelLoaderRegistry,
     ModelLoaderRegistryBase,
 )
 from invokeai.backend.model_manager.load.convert_cache import ModelConvertCacheBase
 from invokeai.backend.model_manager.load.model_cache.model_cache_base import ModelCacheBase
+from invokeai.backend.model_manager.load.model_loaders.generic_diffusers import GenericDiffusersLoader
+from invokeai.backend.util.devices import TorchDevice
 from invokeai.backend.util.logging import InvokeAILogger
 
 from .model_load_base import ModelLoadServiceBase
@@ -39,6 +46,7 @@ class ModelLoadService(ModelLoadServiceBase):
         self._registry = registry
 
     def start(self, invoker: Invoker) -> None:
+        """Start the service."""
         self._invoker = invoker
 
     @property
@@ -46,30 +54,28 @@ class ModelLoadService(ModelLoadServiceBase):
         """Return the RAM cache used by this loader."""
         return self._ram_cache
 
+    @property
+    def gpu_count(self) -> int:
+        """Return the number of GPUs available for our uses."""
+        return len(self._ram_cache.execution_devices)
+
     @property
     def convert_cache(self) -> ModelConvertCacheBase:
         """Return the checkpoint convert cache used by this loader."""
         return self._convert_cache
 
-    def load_model(
-        self,
-        model_config: AnyModelConfig,
-        submodel_type: Optional[SubModelType] = None,
-        context_data: Optional[InvocationContextData] = None,
-    ) -> LoadedModel:
+    def load_model(self, model_config: AnyModelConfig, submodel_type: Optional[SubModelType] = None) -> LoadedModel:
         """
         Given a model's configuration, load it and return the LoadedModel object.
 
         :param model_config: Model configuration record (as returned by ModelRecordBase.get_model())
         :param submodel: For main (pipeline models), the submodel to fetch.
-        :param context: Invocation context used for event reporting
         """
-        if context_data:
-            self._emit_load_event(
-                context_data=context_data,
-                model_config=model_config,
-                submodel_type=submodel_type,
-            )
+
+        # We don't have an invoker during testing
+        # TODO(psyche): Mock this method on the invoker in the tests
+        if hasattr(self, "_invoker"):
+            self._invoker.services.events.emit_model_load_started(model_config, submodel_type)
 
         implementation, model_config, submodel_type = self._registry.get_implementation(model_config, submodel_type)  # type: ignore
         loaded_model: LoadedModel = implementation(
@@ -79,40 +85,45 @@ class ModelLoadService(ModelLoadServiceBase):
             convert_cache=self._convert_cache,
         ).load_model(model_config, submodel_type)
 
-        if context_data:
-            self._emit_load_event(
-                context_data=context_data,
-                model_config=model_config,
-                submodel_type=submodel_type,
-                loaded=True,
-            )
+        if hasattr(self, "_invoker"):
+            self._invoker.services.events.emit_model_load_complete(model_config, submodel_type)
+
         return loaded_model
 
-    def _emit_load_event(
-        self,
-        context_data: InvocationContextData,
-        model_config: AnyModelConfig,
-        loaded: Optional[bool] = False,
-        submodel_type: Optional[SubModelType] = None,
-    ) -> None:
-        if not self._invoker:
-            return
+    def load_model_from_path(
+        self, model_path: Path, loader: Optional[Callable[[Path], AnyModel]] = None
+    ) -> LoadedModelWithoutConfig:
+        cache_key = str(model_path)
+        ram_cache = self.ram_cache
+        try:
+            return LoadedModelWithoutConfig(_locker=ram_cache.get(key=cache_key))
+        except IndexError:
+            pass
 
-        if not loaded:
-            self._invoker.services.events.emit_model_load_started(
-                queue_id=context_data.queue_item.queue_id,
-                queue_item_id=context_data.queue_item.item_id,
-                queue_batch_id=context_data.queue_item.batch_id,
-                graph_execution_state_id=context_data.queue_item.session_id,
-                model_config=model_config,
-                submodel_type=submodel_type,
-            )
-        else:
-            self._invoker.services.events.emit_model_load_completed(
-                queue_id=context_data.queue_item.queue_id,
-                queue_item_id=context_data.queue_item.item_id,
-                queue_batch_id=context_data.queue_item.batch_id,
-                graph_execution_state_id=context_data.queue_item.session_id,
-                model_config=model_config,
-                submodel_type=submodel_type,
-            )
+        def torch_load_file(checkpoint: Path) -> AnyModel:
+            scan_result = scan_file_path(checkpoint)
+            if scan_result.infected_files != 0:
+                raise Exception("The model at {checkpoint} is potentially infected by malware. Aborting load.")
+            result = torch_load(checkpoint, map_location="cpu")
+            return result
+
+        def diffusers_load_directory(directory: Path) -> AnyModel:
+            load_class = GenericDiffusersLoader(
+                app_config=self._app_config,
+                logger=self._logger,
+                ram_cache=self._ram_cache,
+                convert_cache=self.convert_cache,
+            ).get_hf_load_class(directory)
+            return load_class.from_pretrained(model_path, torch_dtype=TorchDevice.choose_torch_dtype())
+
+        loader = loader or (
+            diffusers_load_directory
+            if model_path.is_dir()
+            else torch_load_file
+            if model_path.suffix.endswith((".ckpt", ".pt", ".pth", ".bin"))
+            else lambda path: safetensors_load_file(path, device="cpu")
+        )
+        assert loader is not None
+        raw_model = loader(model_path)
+        ram_cache.put(key=cache_key, model=raw_model)
+        return LoadedModelWithoutConfig(_locker=ram_cache.get(key=cache_key))

invokeai/app/services/model_manager/model_manager_base.py

@@ -1,6 +1,7 @@
 # Copyright (c) 2023 Lincoln D. Stein and the InvokeAI Team
 
 from abc import ABC, abstractmethod
+from typing import Optional, Set
 
 import torch
 from typing_extensions import Self
@@ -31,7 +32,7 @@ class ModelManagerServiceBase(ABC):
         model_record_service: ModelRecordServiceBase,
         download_queue: DownloadQueueServiceBase,
         events: EventServiceBase,
-        execution_device: torch.device,
+        execution_devices: Optional[Set[torch.device]] = None,
     ) -> Self:
         """
         Construct the model manager service instance.

invokeai/app/services/model_manager/model_manager_default.py

@@ -1,14 +1,10 @@
 # Copyright (c) 2023 Lincoln D. Stein and the InvokeAI Team
 """Implementation of ModelManagerServiceBase."""
 
-from typing import Optional
-
-import torch
 from typing_extensions import Self
 
 from invokeai.app.services.invoker import Invoker
 from invokeai.backend.model_manager.load import ModelCache, ModelConvertCache, ModelLoaderRegistry
-from invokeai.backend.util.devices import TorchDevice
 from invokeai.backend.util.logging import InvokeAILogger
 
 from ..config import InvokeAIAppConfig
@@ -69,7 +65,6 @@ class ModelManagerService(ModelManagerServiceBase):
         model_record_service: ModelRecordServiceBase,
         download_queue: DownloadQueueServiceBase,
         events: EventServiceBase,
-        execution_device: Optional[torch.device] = None,
     ) -> Self:
         """
         Construct the model manager service instance.
@@ -82,9 +77,7 @@ class ModelManagerService(ModelManagerServiceBase):
         ram_cache = ModelCache(
             max_cache_size=app_config.ram,
             max_vram_cache_size=app_config.vram,
-            lazy_offloading=app_config.lazy_offload,
             logger=logger,
-            execution_device=execution_device or TorchDevice.choose_torch_device(),
         )
         convert_cache = ModelConvertCache(cache_path=app_config.convert_cache_path, max_size=app_config.convert_cache)
         loader = ModelLoadService(

invokeai/app/services/model_records/model_records_base.py

@@ -12,15 +12,13 @@ from pydantic import BaseModel, Field
 
 from invokeai.app.services.shared.pagination import PaginatedResults
 from invokeai.app.util.model_exclude_null import BaseModelExcludeNull
-from invokeai.backend.model_manager import (
+from invokeai.backend.model_manager.config import (
     AnyModelConfig,
     BaseModelType,
-    ModelFormat,
-    ModelType,
-)
-from invokeai.backend.model_manager.config import (
     ControlAdapterDefaultSettings,
     MainModelDefaultSettings,
+    ModelFormat,
+    ModelType,
     ModelVariantType,
     SchedulerPredictionType,
 )

invokeai/app/services/object_serializer/object_serializer_disk.py

@@ -1,5 +1,6 @@
 import shutil
 import tempfile
+import threading
 import typing
 from pathlib import Path
 from typing import TYPE_CHECKING, Optional, TypeVar
@@ -9,6 +10,7 @@ import torch
 from invokeai.app.services.object_serializer.object_serializer_base import ObjectSerializerBase
 from invokeai.app.services.object_serializer.object_serializer_common import ObjectNotFoundError
 from invokeai.app.util.misc import uuid_string
+from invokeai.backend.util.devices import TorchDevice
 
 if TYPE_CHECKING:
     from invokeai.app.services.invoker import Invoker
@@ -70,7 +72,10 @@ class ObjectSerializerDisk(ObjectSerializerBase[T]):
         return self._output_dir / name
 
     def _new_name(self) -> str:
-        return f"{self._obj_class_name}_{uuid_string()}"
+        tid = threading.current_thread().ident
+        # Add tid to the object name because uuid4 not thread-safe on windows
+        # See https://stackoverflow.com/questions/2759644/python-multiprocessing-doesnt-play-nicely-with-uuid-uuid4
+        return f"{self._obj_class_name}_{tid}-{uuid_string()}"
 
     def _tempdir_cleanup(self) -> None:
         """Calls `cleanup` on the temporary directory, if it exists."""

invokeai/app/services/session_processor/session_processor_base.py

@@ -1,6 +1,49 @@
 from abc import ABC, abstractmethod
+from threading import Event
+from typing import Optional, Protocol
 
+from invokeai.app.invocations.baseinvocation import BaseInvocation, BaseInvocationOutput
+from invokeai.app.services.invocation_services import InvocationServices
 from invokeai.app.services.session_processor.session_processor_common import SessionProcessorStatus
+from invokeai.app.services.session_queue.session_queue_common import SessionQueueItem
+from invokeai.app.util.profiler import Profiler
+
+
+class SessionRunnerBase(ABC):
+    """
+    Base class for session runner.
+    """
+
+    @abstractmethod
+    def start(self, services: InvocationServices, cancel_event: Event, profiler: Optional[Profiler] = None) -> None:
+        """Starts the session runner.
+
+        Args:
+            services: The invocation services.
+            cancel_event: The cancel event.
+            profiler: The profiler to use for session profiling via cProfile. Omit to disable profiling. Basic session
+                stats will be still be recorded and logged when profiling is disabled.
+        """
+        pass
+
+    @abstractmethod
+    def run(self, queue_item: SessionQueueItem) -> None:
+        """Runs a session.
+
+        Args:
+            queue_item: The session to run.
+        """
+        pass
+
+    @abstractmethod
+    def run_node(self, invocation: BaseInvocation, queue_item: SessionQueueItem) -> None:
+        """Run a single node in the graph.
+
+        Args:
+            invocation: The invocation to run.
+            queue_item: The session queue item.
+        """
+        pass
 
 
 class SessionProcessorBase(ABC):
@@ -26,3 +69,85 @@ class SessionProcessorBase(ABC):
     def get_status(self) -> SessionProcessorStatus:
         """Gets the status of the session processor"""
         pass
+
+
+class OnBeforeRunNode(Protocol):
+    def __call__(self, invocation: BaseInvocation, queue_item: SessionQueueItem) -> None:
+        """Callback to run before executing a node.
+
+        Args:
+            invocation: The invocation that will be executed.
+            queue_item: The session queue item.
+        """
+        ...
+
+
+class OnAfterRunNode(Protocol):
+    def __call__(self, invocation: BaseInvocation, queue_item: SessionQueueItem, output: BaseInvocationOutput) -> None:
+        """Callback to run before executing a node.
+
+        Args:
+            invocation: The invocation that was executed.
+            queue_item: The session queue item.
+        """
+        ...
+
+
+class OnNodeError(Protocol):
+    def __call__(
+        self,
+        invocation: BaseInvocation,
+        queue_item: SessionQueueItem,
+        error_type: str,
+        error_message: str,
+        error_traceback: str,
+    ) -> None:
+        """Callback to run when a node has an error.
+
+        Args:
+            invocation: The invocation that errored.
+            queue_item: The session queue item.
+            error_type: The type of error, e.g. "ValueError".
+            error_message: The error message, e.g. "Invalid value".
+            error_traceback: The stringified error traceback.
+        """
+        ...
+
+
+class OnBeforeRunSession(Protocol):
+    def __call__(self, queue_item: SessionQueueItem) -> None:
+        """Callback to run before executing a session.
+
+        Args:
+            queue_item: The session queue item.
+        """
+        ...
+
+
+class OnAfterRunSession(Protocol):
+    def __call__(self, queue_item: SessionQueueItem) -> None:
+        """Callback to run after executing a session.
+
+        Args:
+            queue_item: The session queue item.
+        """
+        ...
+
+
+class OnNonFatalProcessorError(Protocol):
+    def __call__(
+        self,
+        queue_item: Optional[SessionQueueItem],
+        error_type: str,
+        error_message: str,
+        error_traceback: str,
+    ) -> None:
+        """Callback to run when a non-fatal error occurs in the processor.
+
+        Args:
+            queue_item: The session queue item, if one was being executed when the error occurred.
+            error_type: The type of error, e.g. "ValueError".
+            error_message: The error message, e.g. "Invalid value".
+            error_traceback: The stringified error traceback.
+        """
+        ...

invokeai/app/services/session_processor/session_processor_default.py

@@ -1,29 +1,336 @@
 import traceback
 from contextlib import suppress
-from threading import BoundedSemaphore, Thread
+from queue import Queue
+from threading import BoundedSemaphore, Lock, Thread
 from threading import Event as ThreadEvent
-from typing import Optional
+from typing import Optional, Set
 
-from fastapi_events.handlers.local import local_handler
-from fastapi_events.typing import Event as FastAPIEvent
-
-from invokeai.app.invocations.baseinvocation import BaseInvocation
-from invokeai.app.services.events.events_base import EventServiceBase
+from invokeai.app.invocations.baseinvocation import BaseInvocation, BaseInvocationOutput
+from invokeai.app.services.events.events_common import (
+    BatchEnqueuedEvent,
+    FastAPIEvent,
+    QueueClearedEvent,
+    QueueItemStatusChangedEvent,
+    register_events,
+)
 from invokeai.app.services.invocation_stats.invocation_stats_common import GESStatsNotFoundError
+from invokeai.app.services.session_processor.session_processor_base import (
+    OnAfterRunNode,
+    OnAfterRunSession,
+    OnBeforeRunNode,
+    OnBeforeRunSession,
+    OnNodeError,
+    OnNonFatalProcessorError,
+)
 from invokeai.app.services.session_processor.session_processor_common import CanceledException
-from invokeai.app.services.session_queue.session_queue_common import SessionQueueItem
+from invokeai.app.services.session_queue.session_queue_common import SessionQueueItem, SessionQueueItemNotFoundError
+from invokeai.app.services.shared.graph import NodeInputError
 from invokeai.app.services.shared.invocation_context import InvocationContextData, build_invocation_context
 from invokeai.app.util.profiler import Profiler
+from invokeai.backend.util.devices import TorchDevice
 
 from ..invoker import Invoker
-from .session_processor_base import SessionProcessorBase
+from .session_processor_base import InvocationServices, SessionProcessorBase, SessionRunnerBase
 from .session_processor_common import SessionProcessorStatus
 
 
+class DefaultSessionRunner(SessionRunnerBase):
+    """Processes a single session's invocations."""
+
+    def __init__(
+        self,
+        on_before_run_session_callbacks: Optional[list[OnBeforeRunSession]] = None,
+        on_before_run_node_callbacks: Optional[list[OnBeforeRunNode]] = None,
+        on_after_run_node_callbacks: Optional[list[OnAfterRunNode]] = None,
+        on_node_error_callbacks: Optional[list[OnNodeError]] = None,
+        on_after_run_session_callbacks: Optional[list[OnAfterRunSession]] = None,
+    ):
+        """
+        Args:
+            on_before_run_session_callbacks: Callbacks to run before the session starts.
+            on_before_run_node_callbacks: Callbacks to run before each node starts.
+            on_after_run_node_callbacks: Callbacks to run after each node completes.
+            on_node_error_callbacks: Callbacks to run when a node errors.
+            on_after_run_session_callbacks: Callbacks to run after the session completes.
+        """
+
+        self._on_before_run_session_callbacks = on_before_run_session_callbacks or []
+        self._on_before_run_node_callbacks = on_before_run_node_callbacks or []
+        self._on_after_run_node_callbacks = on_after_run_node_callbacks or []
+        self._on_node_error_callbacks = on_node_error_callbacks or []
+        self._on_after_run_session_callbacks = on_after_run_session_callbacks or []
+        self._process_lock = Lock()
+
+    def start(
+        self, services: InvocationServices, cancel_event: ThreadEvent, profiler: Optional[Profiler] = None
+    ) -> None:
+        self._services = services
+        self._cancel_event = cancel_event
+        self._profiler = profiler
+
+    def _is_canceled(self) -> bool:
+        """Check if the cancel event is set. This is also passed to the invocation context builder and called during
+        denoising to check if the session has been canceled."""
+        return self._cancel_event.is_set()
+
+    def run(self, queue_item: SessionQueueItem):
+        # Exceptions raised outside `run_node` are handled by the processor. There is no need to catch them here.
+
+        self._on_before_run_session(queue_item=queue_item)
+
+        # Loop over invocations until the session is complete or canceled
+        while True:
+            try:
+                with self._process_lock:
+                    invocation = queue_item.session.next()
+            # Anything other than a `NodeInputError` is handled as a processor error
+            except NodeInputError as e:
+                error_type = e.__class__.__name__
+                error_message = str(e)
+                error_traceback = traceback.format_exc()
+                self._on_node_error(
+                    invocation=e.node,
+                    queue_item=queue_item,
+                    error_type=error_type,
+                    error_message=error_message,
+                    error_traceback=error_traceback,
+                )
+                break
+
+            if invocation is None or self._is_canceled():
+                break
+
+            self.run_node(invocation, queue_item)
+
+            # The session is complete if all invocations have been run or there is an error on the session.
+            # At this time, the queue item may be canceled, but the object itself here won't be updated yet. We must
+            # use the cancel event to check if the session is canceled.
+            if (
+                queue_item.session.is_complete()
+                or self._is_canceled()
+                or queue_item.status in ["failed", "canceled", "completed"]
+            ):
+                break
+
+        self._on_after_run_session(queue_item=queue_item)
+
+    def run_node(self, invocation: BaseInvocation, queue_item: SessionQueueItem) -> None:
+        try:
+            # Any unhandled exception in this scope is an invocation error & will fail the graph
+            with self._services.performance_statistics.collect_stats(invocation, queue_item.session_id):
+                self._on_before_run_node(invocation, queue_item)
+
+                data = InvocationContextData(
+                    invocation=invocation,
+                    source_invocation_id=queue_item.session.prepared_source_mapping[invocation.id],
+                    queue_item=queue_item,
+                )
+                context = build_invocation_context(
+                    data=data,
+                    services=self._services,
+                    is_canceled=self._is_canceled,
+                )
+
+                # Invoke the node
+                output = invocation.invoke_internal(context=context, services=self._services)
+                # Save output and history
+                queue_item.session.complete(invocation.id, output)
+
+                self._on_after_run_node(invocation, queue_item, output)
+
+        except KeyboardInterrupt:
+            # TODO(psyche): This is expected to be caught in the main thread. Do we need to catch this here?
+            pass
+        except CanceledException:
+            # A CanceledException is raised during the denoising step callback if the cancel event is set. We don't need
+            # to do any handling here, and no error should be set - just pass and the cancellation will be handled
+            # correctly in the next iteration of the session runner loop.
+            #
+            # See the comment in the processor's `_on_queue_item_status_changed()` method for more details on how we
+            # handle cancellation.
+            pass
+        except Exception as e:
+            error_type = e.__class__.__name__
+            error_message = str(e)
+            error_traceback = traceback.format_exc()
+            self._on_node_error(
+                invocation=invocation,
+                queue_item=queue_item,
+                error_type=error_type,
+                error_message=error_message,
+                error_traceback=error_traceback,
+            )
+
+    def _on_before_run_session(self, queue_item: SessionQueueItem) -> None:
+        """Called before a session is run.
+
+        - Start the profiler if profiling is enabled.
+        - Run any callbacks registered for this event.
+        """
+
+        self._services.logger.debug(
+            f"On before run session: queue item {queue_item.item_id}, session {queue_item.session_id}"
+        )
+
+        # If profiling is enabled, start the profiler
+        if self._profiler is not None:
+            self._profiler.start(profile_id=queue_item.session_id)
+
+        for callback in self._on_before_run_session_callbacks:
+            callback(queue_item=queue_item)
+
+    def _on_after_run_session(self, queue_item: SessionQueueItem) -> None:
+        """Called after a session is run.
+
+        - Stop the profiler if profiling is enabled.
+        - Update the queue item's session object in the database.
+        - If not already canceled or failed, complete the queue item.
+        - Log and reset performance statistics.
+        - Run any callbacks registered for this event.
+        """
+
+        self._services.logger.debug(
+            f"On after run session: queue item {queue_item.item_id}, session {queue_item.session_id}"
+        )
+
+        # If we are profiling, stop the profiler and dump the profile & stats
+        if self._profiler is not None:
+            profile_path = self._profiler.stop()
+            stats_path = profile_path.with_suffix(".json")
+            self._services.performance_statistics.dump_stats(
+                graph_execution_state_id=queue_item.session.id, output_path=stats_path
+            )
+
+        try:
+            # Update the queue item with the completed session. If the queue item has been removed from the queue,
+            # we'll get a SessionQueueItemNotFoundError and we can ignore it. This can happen if the queue is cleared
+            # while the session is running.
+            queue_item = self._services.session_queue.set_queue_item_session(queue_item.item_id, queue_item.session)
+
+            # The queue item may have been canceled or failed while the session was running. We should only complete it
+            # if it is not already canceled or failed.
+            if queue_item.status not in ["canceled", "failed"]:
+                queue_item = self._services.session_queue.complete_queue_item(queue_item.item_id)
+
+            # We'll get a GESStatsNotFoundError if we try to log stats for an untracked graph, but in the processor
+            # we don't care about that - suppress the error.
+            with suppress(GESStatsNotFoundError):
+                self._services.performance_statistics.log_stats(queue_item.session.id)
+                self._services.performance_statistics.reset_stats(queue_item.session.id)
+
+            for callback in self._on_after_run_session_callbacks:
+                callback(queue_item=queue_item)
+        except SessionQueueItemNotFoundError:
+            pass
+
+    def _on_before_run_node(self, invocation: BaseInvocation, queue_item: SessionQueueItem):
+        """Called before a node is run.
+
+        - Emits an invocation started event.
+        - Run any callbacks registered for this event.
+        """
+
+        self._services.logger.debug(
+            f"On before run node: queue item {queue_item.item_id}, session {queue_item.session_id}, node {invocation.id} ({invocation.get_type()})"
+        )
+
+        # Send starting event
+        self._services.events.emit_invocation_started(queue_item=queue_item, invocation=invocation)
+
+        for callback in self._on_before_run_node_callbacks:
+            callback(invocation=invocation, queue_item=queue_item)
+
+    def _on_after_run_node(
+        self, invocation: BaseInvocation, queue_item: SessionQueueItem, output: BaseInvocationOutput
+    ):
+        """Called after a node is run.
+
+        - Emits an invocation complete event.
+        - Run any callbacks registered for this event.
+        """
+
+        self._services.logger.debug(
+            f"On after run node: queue item {queue_item.item_id}, session {queue_item.session_id}, node {invocation.id} ({invocation.get_type()})"
+        )
+
+        # Send complete event on successful runs
+        self._services.events.emit_invocation_complete(invocation=invocation, queue_item=queue_item, output=output)
+
+        for callback in self._on_after_run_node_callbacks:
+            callback(invocation=invocation, queue_item=queue_item, output=output)
+
+    def _on_node_error(
+        self,
+        invocation: BaseInvocation,
+        queue_item: SessionQueueItem,
+        error_type: str,
+        error_message: str,
+        error_traceback: str,
+    ):
+        """Called when a node errors. Node errors may occur when running or preparing the node..
+
+        - Set the node error on the session object.
+        - Log the error.
+        - Fail the queue item.
+        - Emits an invocation error event.
+        - Run any callbacks registered for this event.
+        """
+
+        self._services.logger.debug(
+            f"On node error: queue item {queue_item.item_id}, session {queue_item.session_id}, node {invocation.id} ({invocation.get_type()})"
+        )
+
+        # Node errors do not get the full traceback. Only the queue item gets the full traceback.
+        node_error = f"{error_type}: {error_message}"
+        queue_item.session.set_node_error(invocation.id, node_error)
+        self._services.logger.error(
+            f"Error while invoking session {queue_item.session_id}, invocation {invocation.id} ({invocation.get_type()}): {error_message}"
+        )
+        self._services.logger.error(error_traceback)
+
+        # Fail the queue item
+        queue_item = self._services.session_queue.set_queue_item_session(queue_item.item_id, queue_item.session)
+        queue_item = self._services.session_queue.fail_queue_item(
+            queue_item.item_id, error_type, error_message, error_traceback
+        )
+
+        # Send error event
+        self._services.events.emit_invocation_error(
+            queue_item=queue_item,
+            invocation=invocation,
+            error_type=error_type,
+            error_message=error_message,
+            error_traceback=error_traceback,
+        )
+
+        for callback in self._on_node_error_callbacks:
+            callback(
+                invocation=invocation,
+                queue_item=queue_item,
+                error_type=error_type,
+                error_message=error_message,
+                error_traceback=error_traceback,
+            )
+
+
 class DefaultSessionProcessor(SessionProcessorBase):
-    def start(self, invoker: Invoker, thread_limit: int = 1, polling_interval: int = 1) -> None:
+    def __init__(
+        self,
+        session_runner: Optional[SessionRunnerBase] = None,
+        on_non_fatal_processor_error_callbacks: Optional[list[OnNonFatalProcessorError]] = None,
+        thread_limit: int = 1,
+        polling_interval: int = 1,
+    ) -> None:
+        super().__init__()
+
+        self.session_runner = session_runner if session_runner else DefaultSessionRunner()
+        self._on_non_fatal_processor_error_callbacks = on_non_fatal_processor_error_callbacks or []
+        self._thread_limit = thread_limit
+        self._polling_interval = polling_interval
+
+    def start(self, invoker: Invoker) -> None:
         self._invoker: Invoker = invoker
-        self._queue_item: Optional[SessionQueueItem] = None
+        self._active_queue_items: Set[SessionQueueItem] = set()
         self._invocation: Optional[BaseInvocation] = None
 
         self._resume_event = ThreadEvent()
@@ -31,11 +338,11 @@ class DefaultSessionProcessor(SessionProcessorBase):
         self._poll_now_event = ThreadEvent()
         self._cancel_event = ThreadEvent()
 
-        local_handler.register(event_name=EventServiceBase.queue_event, _func=self._on_queue_event)
+        register_events(QueueClearedEvent, self._on_queue_cleared)
+        register_events(BatchEnqueuedEvent, self._on_batch_enqueued)
+        register_events(QueueItemStatusChangedEvent, self._on_queue_item_status_changed)
 
-        self._thread_limit = thread_limit
-        self._thread_semaphore = BoundedSemaphore(thread_limit)
-        self._polling_interval = polling_interval
+        self._thread_semaphore = BoundedSemaphore(self._thread_limit)
 
         # If profiling is enabled, create a profiler. The same profiler will be used for all sessions. Internally,
         # the profiler will create a new profile for each session.
@@ -49,6 +356,14 @@ class DefaultSessionProcessor(SessionProcessorBase):
             else None
         )
 
+        self._worker_thread_count = self._invoker.services.configuration.max_threads or len(
+            TorchDevice.execution_devices()
+        )
+
+        self._session_worker_queue: Queue[SessionQueueItem] = Queue()
+
+        self.session_runner.start(services=invoker.services, cancel_event=self._cancel_event, profiler=self._profiler)
+        # Session processor - singlethreaded
         self._thread = Thread(
             name="session_processor",
             target=self._process,
@@ -61,36 +376,41 @@ class DefaultSessionProcessor(SessionProcessorBase):
         )
         self._thread.start()
 
+        # Session processor workers - multithreaded
+        self._invoker.services.logger.debug(f"Starting {self._worker_thread_count} session processing threads.")
+        for _i in range(0, self._worker_thread_count):
+            worker = Thread(
+                name="session_worker",
+                target=self._process_next_session,
+                daemon=True,
+            )
+            worker.start()
+
     def stop(self, *args, **kwargs) -> None:
         self._stop_event.set()
 
     def _poll_now(self) -> None:
         self._poll_now_event.set()
 
-    async def _on_queue_event(self, event: FastAPIEvent) -> None:
-        event_name = event[1]["event"]
+    async def _on_queue_cleared(self, event: FastAPIEvent[QueueClearedEvent]) -> None:
+        if any(item.queue_id == event[1].queue_id for item in self._active_queue_items):
+            self._cancel_event.set()
+            self._poll_now()
 
-        if (
-            event_name == "session_canceled"
-            and self._queue_item
-            and self._queue_item.item_id == event[1]["data"]["queue_item_id"]
-        ):
-            self._cancel_event.set()
-            self._poll_now()
-        elif (
-            event_name == "queue_cleared"
-            and self._queue_item
-            and self._queue_item.queue_id == event[1]["data"]["queue_id"]
-        ):
-            self._cancel_event.set()
-            self._poll_now()
-        elif event_name == "batch_enqueued":
-            self._poll_now()
-        elif event_name == "queue_item_status_changed" and event[1]["data"]["queue_item"]["status"] in [
-            "completed",
-            "failed",
-            "canceled",
-        ]:
+    async def _on_batch_enqueued(self, event: FastAPIEvent[BatchEnqueuedEvent]) -> None:
+        self._poll_now()
+
+    async def _on_queue_item_status_changed(self, event: FastAPIEvent[QueueItemStatusChangedEvent]) -> None:
+        if self._active_queue_items and event[1].status in ["completed", "failed", "canceled"]:
+            # When the queue item is canceled via HTTP, the queue item status is set to `"canceled"` and this event is
+            # emitted. We need to respond to this event and stop graph execution. This is done by setting the cancel
+            # event, which the session runner checks between invocations. If set, the session runner loop is broken.
+            #
+            # Long-running nodes that cannot be interrupted easily present a challenge. `denoise_latents` is one such
+            # node, but it gets a step callback, called on each step of denoising. This callback checks if the queue item
+            # is canceled, and if it is, raises a `CanceledException` to stop execution immediately.
+            if event[1].status == "canceled":
+                self._cancel_event.set()
             self._poll_now()
 
     def resume(self) -> SessionProcessorStatus:
@@ -106,7 +426,7 @@ class DefaultSessionProcessor(SessionProcessorBase):
     def get_status(self) -> SessionProcessorStatus:
         return SessionProcessorStatus(
             is_started=self._resume_event.is_set(),
-            is_processing=self._queue_item is not None,
+            is_processing=len(self._active_queue_items) > 0,
         )
 
     def _process(
@@ -116,8 +436,8 @@ class DefaultSessionProcessor(SessionProcessorBase):
         resume_event: ThreadEvent,
         cancel_event: ThreadEvent,
     ):
-        # Outermost processor try block; any unhandled exception is a fatal processor error
         try:
+            # Any unhandled exception in this block is a fatal processor error and will stop the processor.
             self._thread_semaphore.acquire()
             stop_event.clear()
             resume_event.set()
@@ -125,174 +445,94 @@ class DefaultSessionProcessor(SessionProcessorBase):
 
             while not stop_event.is_set():
                 poll_now_event.clear()
-                # Middle processor try block; any unhandled exception is a non-fatal processor error
                 try:
+                    # Any unhandled exception in this block is a nonfatal processor error and will be handled.
                     # If we are paused, wait for resume event
                     resume_event.wait()
 
                     # Get the next session to process
-                    self._queue_item = self._invoker.services.session_queue.dequeue()
+                    queue_item = self._invoker.services.session_queue.dequeue()
 
-                    if self._queue_item is None:
+                    if queue_item is None:
                         # The queue was empty, wait for next polling interval or event to try again
                         self._invoker.services.logger.debug("Waiting for next polling interval or event")
                         poll_now_event.wait(self._polling_interval)
                         continue
 
-                    self._invoker.services.logger.debug(f"Executing queue item {self._queue_item.item_id}")
+                    self._session_worker_queue.put(queue_item)
+                    self._invoker.services.logger.debug(f"Scheduling queue item {queue_item.item_id} to run")
                     cancel_event.clear()
 
-                    # If profiling is enabled, start the profiler
-                    if self._profiler is not None:
-                        self._profiler.start(profile_id=self._queue_item.session_id)
+                    # Run the graph
+                    # self.session_runner.run(queue_item=self._queue_item)
 
-                    # Prepare invocations and take the first
-                    self._invocation = self._queue_item.session.next()
-
-                    # Loop over invocations until the session is complete or canceled
-                    while self._invocation is not None and not cancel_event.is_set():
-                        # get the source node id to provide to clients (the prepared node id is not as useful)
-                        source_invocation_id = self._queue_item.session.prepared_source_mapping[self._invocation.id]
-
-                        # Send starting event
-                        self._invoker.services.events.emit_invocation_started(
-                            queue_batch_id=self._queue_item.batch_id,
-                            queue_item_id=self._queue_item.item_id,
-                            queue_id=self._queue_item.queue_id,
-                            graph_execution_state_id=self._queue_item.session_id,
-                            node=self._invocation.model_dump(),
-                            source_node_id=source_invocation_id,
-                        )
-
-                        # Innermost processor try block; any unhandled exception is an invocation error & will fail the graph
-                        try:
-                            with self._invoker.services.performance_statistics.collect_stats(
-                                self._invocation, self._queue_item.session.id
-                            ):
-                                # Build invocation context (the node-facing API)
-                                data = InvocationContextData(
-                                    invocation=self._invocation,
-                                    source_invocation_id=source_invocation_id,
-                                    queue_item=self._queue_item,
-                                )
-                                context = build_invocation_context(
-                                    data=data,
-                                    services=self._invoker.services,
-                                    cancel_event=self._cancel_event,
-                                )
-
-                                # Invoke the node
-                                outputs = self._invocation.invoke_internal(
-                                    context=context, services=self._invoker.services
-                                )
-
-                                # Save outputs and history
-                                self._queue_item.session.complete(self._invocation.id, outputs)
-
-                                # Send complete event
-                                self._invoker.services.events.emit_invocation_complete(
-                                    queue_batch_id=self._queue_item.batch_id,
-                                    queue_item_id=self._queue_item.item_id,
-                                    queue_id=self._queue_item.queue_id,
-                                    graph_execution_state_id=self._queue_item.session.id,
-                                    node=self._invocation.model_dump(),
-                                    source_node_id=source_invocation_id,
-                                    result=outputs.model_dump(),
-                                )
-
-                        except KeyboardInterrupt:
-                            # TODO(MM2): Create an event for this
-                            pass
-
-                        except CanceledException:
-                            # When the user cancels the graph, we first set the cancel event. The event is checked
-                            # between invocations, in this loop. Some invocations are long-running, and we need to
-                            # be able to cancel them mid-execution.
-                            #
-                            # For example, denoising is a long-running invocation with many steps. A step callback
-                            # is executed after each step. This step callback checks if the canceled event is set,
-                            # then raises a CanceledException to stop execution immediately.
-                            #
-                            # When we get a CanceledException, we don't need to do anything - just pass and let the
-                            # loop go to its next iteration, and the cancel event will be handled correctly.
-                            pass
-
-                        except Exception as e:
-                            error = traceback.format_exc()
-
-                            # Save error
-                            self._queue_item.session.set_node_error(self._invocation.id, error)
-                            self._invoker.services.logger.error(
-                                f"Error while invoking session {self._queue_item.session_id}, invocation {self._invocation.id} ({self._invocation.get_type()}):\n{e}"
-                            )
-                            self._invoker.services.logger.error(error)
-
-                            # Send error event
-                            self._invoker.services.events.emit_invocation_error(
-                                queue_batch_id=self._queue_item.session_id,
-                                queue_item_id=self._queue_item.item_id,
-                                queue_id=self._queue_item.queue_id,
-                                graph_execution_state_id=self._queue_item.session.id,
-                                node=self._invocation.model_dump(),
-                                source_node_id=source_invocation_id,
-                                error_type=e.__class__.__name__,
-                                error=error,
-                            )
-                            pass
-
-                        # The session is complete if the all invocations are complete or there was an error
-                        if self._queue_item.session.is_complete() or cancel_event.is_set():
-                            # Send complete event
-                            self._invoker.services.events.emit_graph_execution_complete(
-                                queue_batch_id=self._queue_item.batch_id,
-                                queue_item_id=self._queue_item.item_id,
-                                queue_id=self._queue_item.queue_id,
-                                graph_execution_state_id=self._queue_item.session.id,
-                            )
-                            # If we are profiling, stop the profiler and dump the profile & stats
-                            if self._profiler:
-                                profile_path = self._profiler.stop()
-                                stats_path = profile_path.with_suffix(".json")
-                                self._invoker.services.performance_statistics.dump_stats(
-                                    graph_execution_state_id=self._queue_item.session.id, output_path=stats_path
-                                )
-                            # We'll get a GESStatsNotFoundError if we try to log stats for an untracked graph, but in the processor
-                            # we don't care about that - suppress the error.
-                            with suppress(GESStatsNotFoundError):
-                                self._invoker.services.performance_statistics.log_stats(self._queue_item.session.id)
-                                self._invoker.services.performance_statistics.reset_stats()
-
-                            # Set the invocation to None to prepare for the next session
-                            self._invocation = None
-                        else:
-                            # Prepare the next invocation
-                            self._invocation = self._queue_item.session.next()
-                    else:
-                        # The queue was empty, wait for next polling interval or event to try again
-                        self._invoker.services.logger.debug("Waiting for next polling interval or event")
-                        poll_now_event.wait(self._polling_interval)
-                        continue
                 except Exception:
-                    # Non-fatal error in processor
-                    self._invoker.services.logger.error(
-                        f"Non-fatal error in session processor:\n{traceback.format_exc()}"
-                    )
-                    # Cancel the queue item
-                    if self._queue_item is not None:
-                        self._invoker.services.session_queue.cancel_queue_item(
-                            self._queue_item.item_id, error=traceback.format_exc()
-                        )
-                    # Reset the invocation to None to prepare for the next session
-                    self._invocation = None
-                    # Immediately poll for next queue item
+                    # Wait for next polling interval or event to try again
                     poll_now_event.wait(self._polling_interval)
                     continue
-        except Exception:
+        except Exception as e:
             # Fatal error in processor, log and pass - we're done here
-            self._invoker.services.logger.error(f"Fatal Error in session processor:\n{traceback.format_exc()}")
+            error_type = e.__class__.__name__
+            error_message = str(e)
+            error_traceback = traceback.format_exc()
+            self._invoker.services.logger.error(f"Fatal Error in session processor {error_type}: {error_message}")
+            self._invoker.services.logger.error(error_traceback)
             pass
         finally:
             stop_event.clear()
             poll_now_event.clear()
-            self._queue_item = None
             self._thread_semaphore.release()
+
+    def _process_next_session(self) -> None:
+        while True:
+            self._resume_event.wait()
+            queue_item = self._session_worker_queue.get()
+            if queue_item.status == "canceled":
+                continue
+            try:
+                self._active_queue_items.add(queue_item)
+                # reserve a GPU for this session - may block
+                with self._invoker.services.model_manager.load.ram_cache.reserve_execution_device():
+                    # Run the session on the reserved GPU
+                    self.session_runner.run(queue_item=queue_item)
+            except Exception:
+                continue
+            finally:
+                self._active_queue_items.remove(queue_item)
+
+    def _on_non_fatal_processor_error(
+        self,
+        queue_item: Optional[SessionQueueItem],
+        error_type: str,
+        error_message: str,
+        error_traceback: str,
+    ) -> None:
+        """Called when a non-fatal error occurs in the processor.
+
+        - Log the error.
+        - If a queue item is provided, update the queue item with the completed session & fail it.
+        - Run any callbacks registered for this event.
+        """
+
+        self._invoker.services.logger.error(f"Non-fatal error in session processor {error_type}: {error_message}")
+        self._invoker.services.logger.error(error_traceback)
+
+        if queue_item is not None:
+            # Update the queue item with the completed session & fail it
+            queue_item = self._invoker.services.session_queue.set_queue_item_session(
+                queue_item.item_id, queue_item.session
+            )
+            queue_item = self._invoker.services.session_queue.fail_queue_item(
+                item_id=queue_item.item_id,
+                error_type=error_type,
+                error_message=error_message,
+                error_traceback=error_traceback,
+            )
+
+        for callback in self._on_non_fatal_processor_error_callbacks:
+            callback(
+                queue_item=queue_item,
+                error_type=error_type,
+                error_message=error_message,
+                error_traceback=error_traceback,
+            )

invokeai/app/services/session_queue/session_queue_base.py

@@ -16,6 +16,7 @@ from invokeai.app.services.session_queue.session_queue_common import (
     SessionQueueItemDTO,
     SessionQueueStatus,
 )
+from invokeai.app.services.shared.graph import GraphExecutionState
 from invokeai.app.services.shared.pagination import CursorPaginatedResults
 
 
@@ -73,10 +74,22 @@ class SessionQueueBase(ABC):
         pass
 
     @abstractmethod
-    def cancel_queue_item(self, item_id: int, error: Optional[str] = None) -> SessionQueueItem:
+    def complete_queue_item(self, item_id: int) -> SessionQueueItem:
+        """Completes a session queue item"""
+        pass
+
+    @abstractmethod
+    def cancel_queue_item(self, item_id: int) -> SessionQueueItem:
         """Cancels a session queue item"""
         pass
 
+    @abstractmethod
+    def fail_queue_item(
+        self, item_id: int, error_type: str, error_message: str, error_traceback: str
+    ) -> SessionQueueItem:
+        """Fails a session queue item"""
+        pass
+
     @abstractmethod
     def cancel_by_batch_ids(self, queue_id: str, batch_ids: list[str]) -> CancelByBatchIDsResult:
         """Cancels all queue items with matching batch IDs"""
@@ -103,3 +116,8 @@ class SessionQueueBase(ABC):
     def get_queue_item(self, item_id: int) -> SessionQueueItem:
         """Gets a session queue item by ID"""
         pass
+
+    @abstractmethod
+    def set_queue_item_session(self, item_id: int, session: GraphExecutionState) -> SessionQueueItem:
+        """Sets the session for a session queue item. Use this to update the session state."""
+        pass

invokeai/app/services/session_queue/session_queue_common.py

@@ -3,7 +3,16 @@ import json
 from itertools import chain, product
 from typing import Generator, Iterable, Literal, NamedTuple, Optional, TypeAlias, Union, cast
 
-from pydantic import BaseModel, ConfigDict, Field, StrictStr, TypeAdapter, field_validator, model_validator
+from pydantic import (
+    AliasChoices,
+    BaseModel,
+    ConfigDict,
+    Field,
+    StrictStr,
+    TypeAdapter,
+    field_validator,
+    model_validator,
+)
 from pydantic_core import to_jsonable_python
 
 from invokeai.app.invocations.baseinvocation import BaseInvocation
@@ -189,7 +198,13 @@ class SessionQueueItemWithoutGraph(BaseModel):
     session_id: str = Field(
         description="The ID of the session associated with this queue item. The session doesn't exist in graph_executions until the queue item is executed."
     )
-    error: Optional[str] = Field(default=None, description="The error message if this queue item errored")
+    error_type: Optional[str] = Field(default=None, description="The error type if this queue item errored")
+    error_message: Optional[str] = Field(default=None, description="The error message if this queue item errored")
+    error_traceback: Optional[str] = Field(
+        default=None,
+        description="The error traceback if this queue item errored",
+        validation_alias=AliasChoices("error_traceback", "error"),
+    )
     created_at: Union[datetime.datetime, str] = Field(description="When this queue item was created")
     updated_at: Union[datetime.datetime, str] = Field(description="When this queue item was updated")
     started_at: Optional[Union[datetime.datetime, str]] = Field(description="When this queue item was started")
@@ -221,6 +236,9 @@ class SessionQueueItemWithoutGraph(BaseModel):
         }
     )
 
+    def __hash__(self) -> int:
+        return self.item_id
+
 
 class SessionQueueItemDTO(SessionQueueItemWithoutGraph):
     pass

invokeai/app/services/session_queue/session_queue_sqlite.py

@@ -2,10 +2,6 @@ import sqlite3
 import threading
 from typing import Optional, Union, cast
 
-from fastapi_events.handlers.local import local_handler
-from fastapi_events.typing import Event as FastAPIEvent
-
-from invokeai.app.services.events.events_base import EventServiceBase
 from invokeai.app.services.invoker import Invoker
 from invokeai.app.services.session_queue.session_queue_base import SessionQueueBase
 from invokeai.app.services.session_queue.session_queue_common import (
@@ -27,6 +23,7 @@ from invokeai.app.services.session_queue.session_queue_common import (
     calc_session_count,
     prepare_values_to_insert,
 )
+from invokeai.app.services.shared.graph import GraphExecutionState
 from invokeai.app.services.shared.pagination import CursorPaginatedResults
 from invokeai.app.services.shared.sqlite.sqlite_database import SqliteDatabase
 
@@ -40,10 +37,14 @@ class SqliteSessionQueue(SessionQueueBase):
     def start(self, invoker: Invoker) -> None:
         self.__invoker = invoker
         self._set_in_progress_to_canceled()
-        prune_result = self.prune(DEFAULT_QUEUE_ID)
-        local_handler.register(event_name=EventServiceBase.queue_event, _func=self._on_session_event)
-        if prune_result.deleted > 0:
-            self.__invoker.services.logger.info(f"Pruned {prune_result.deleted} finished queue items")
+        if self.__invoker.services.configuration.clear_queue_on_startup:
+            clear_result = self.clear(DEFAULT_QUEUE_ID)
+            if clear_result.deleted > 0:
+                self.__invoker.services.logger.info(f"Cleared all {clear_result.deleted} queue items")
+        else:
+            prune_result = self.prune(DEFAULT_QUEUE_ID)
+            if prune_result.deleted > 0:
+                self.__invoker.services.logger.info(f"Pruned {prune_result.deleted} finished queue items")
 
     def __init__(self, db: SqliteDatabase) -> None:
         super().__init__()
@@ -51,52 +52,6 @@ class SqliteSessionQueue(SessionQueueBase):
         self.__conn = db.conn
         self.__cursor = self.__conn.cursor()
 
-    def _match_event_name(self, event: FastAPIEvent, match_in: list[str]) -> bool:
-        return event[1]["event"] in match_in
-
-    async def _on_session_event(self, event: FastAPIEvent) -> FastAPIEvent:
-        event_name = event[1]["event"]
-
-        # This was a match statement, but match is not supported on python 3.9
-        if event_name == "graph_execution_state_complete":
-            await self._handle_complete_event(event)
-        elif event_name == "invocation_error":
-            await self._handle_error_event(event)
-        elif event_name == "session_canceled":
-            await self._handle_cancel_event(event)
-        return event
-
-    async def _handle_complete_event(self, event: FastAPIEvent) -> None:
-        try:
-            item_id = event[1]["data"]["queue_item_id"]
-            # When a queue item has an error, we get an error event, then a completed event.
-            # Mark the queue item completed only if it isn't already marked completed, e.g.
-            # by a previously-handled error event.
-            queue_item = self.get_queue_item(item_id)
-            if queue_item.status not in ["completed", "failed", "canceled"]:
-                queue_item = self._set_queue_item_status(item_id=queue_item.item_id, status="completed")
-        except SessionQueueItemNotFoundError:
-            return
-
-    async def _handle_error_event(self, event: FastAPIEvent) -> None:
-        try:
-            item_id = event[1]["data"]["queue_item_id"]
-            error = event[1]["data"]["error"]
-            queue_item = self.get_queue_item(item_id)
-            # always set to failed if have an error, even if previously the item was marked completed or canceled
-            queue_item = self._set_queue_item_status(item_id=queue_item.item_id, status="failed", error=error)
-        except SessionQueueItemNotFoundError:
-            return
-
-    async def _handle_cancel_event(self, event: FastAPIEvent) -> None:
-        try:
-            item_id = event[1]["data"]["queue_item_id"]
-            queue_item = self.get_queue_item(item_id)
-            if queue_item.status not in ["completed", "failed", "canceled"]:
-                queue_item = self._set_queue_item_status(item_id=queue_item.item_id, status="canceled")
-        except SessionQueueItemNotFoundError:
-            return
-
     def _set_in_progress_to_canceled(self) -> None:
         """
         Sets all in_progress queue items to canceled. Run on app startup, not associated with any queue.
@@ -271,17 +226,22 @@ class SqliteSessionQueue(SessionQueueBase):
         return SessionQueueItem.queue_item_from_dict(dict(result))
 
     def _set_queue_item_status(
-        self, item_id: int, status: QUEUE_ITEM_STATUS, error: Optional[str] = None
+        self,
+        item_id: int,
+        status: QUEUE_ITEM_STATUS,
+        error_type: Optional[str] = None,
+        error_message: Optional[str] = None,
+        error_traceback: Optional[str] = None,
     ) -> SessionQueueItem:
         try:
             self.__lock.acquire()
             self.__cursor.execute(
                 """--sql
                 UPDATE session_queue
-                SET status = ?, error = ?
+                SET status = ?, error_type = ?, error_message = ?, error_traceback = ?
                 WHERE item_id = ?
                 """,
-                (status, error, item_id),
+                (status, error_type, error_message, error_traceback, item_id),
             )
             self.__conn.commit()
         except Exception:
@@ -292,11 +252,7 @@ class SqliteSessionQueue(SessionQueueBase):
         queue_item = self.get_queue_item(item_id)
         batch_status = self.get_batch_status(queue_id=queue_item.queue_id, batch_id=queue_item.batch_id)
         queue_status = self.get_queue_status(queue_id=queue_item.queue_id)
-        self.__invoker.services.events.emit_queue_item_status_changed(
-            session_queue_item=queue_item,
-            batch_status=batch_status,
-            queue_status=queue_status,
-        )
+        self.__invoker.services.events.emit_queue_item_status_changed(queue_item, batch_status, queue_status)
         return queue_item
 
     def is_empty(self, queue_id: str) -> IsEmptyResult:
@@ -338,26 +294,6 @@ class SqliteSessionQueue(SessionQueueBase):
             self.__lock.release()
         return IsFullResult(is_full=is_full)
 
-    def delete_queue_item(self, item_id: int) -> SessionQueueItem:
-        queue_item = self.get_queue_item(item_id=item_id)
-        try:
-            self.__lock.acquire()
-            self.__cursor.execute(
-                """--sql
-                DELETE FROM session_queue
-                WHERE
-                  item_id = ?
-                """,
-                (item_id,),
-            )
-            self.__conn.commit()
-        except Exception:
-            self.__conn.rollback()
-            raise
-        finally:
-            self.__lock.release()
-        return queue_item
-
     def clear(self, queue_id: str) -> ClearResult:
         try:
             self.__lock.acquire()
@@ -424,17 +360,28 @@ class SqliteSessionQueue(SessionQueueBase):
             self.__lock.release()
         return PruneResult(deleted=count)
 
-    def cancel_queue_item(self, item_id: int, error: Optional[str] = None) -> SessionQueueItem:
-        queue_item = self.get_queue_item(item_id)
-        if queue_item.status not in ["canceled", "failed", "completed"]:
-            status = "failed" if error is not None else "canceled"
-            queue_item = self._set_queue_item_status(item_id=item_id, status=status, error=error)  # type: ignore [arg-type] # mypy seems to not narrow the Literals here
-            self.__invoker.services.events.emit_session_canceled(
-                queue_item_id=queue_item.item_id,
-                queue_id=queue_item.queue_id,
-                queue_batch_id=queue_item.batch_id,
-                graph_execution_state_id=queue_item.session_id,
-            )
+    def cancel_queue_item(self, item_id: int) -> SessionQueueItem:
+        queue_item = self._set_queue_item_status(item_id=item_id, status="canceled")
+        return queue_item
+
+    def complete_queue_item(self, item_id: int) -> SessionQueueItem:
+        queue_item = self._set_queue_item_status(item_id=item_id, status="completed")
+        return queue_item
+
+    def fail_queue_item(
+        self,
+        item_id: int,
+        error_type: str,
+        error_message: str,
+        error_traceback: str,
+    ) -> SessionQueueItem:
+        queue_item = self._set_queue_item_status(
+            item_id=item_id,
+            status="failed",
+            error_type=error_type,
+            error_message=error_message,
+            error_traceback=error_traceback,
+        )
         return queue_item
 
     def cancel_by_batch_ids(self, queue_id: str, batch_ids: list[str]) -> CancelByBatchIDsResult:
@@ -470,18 +417,10 @@ class SqliteSessionQueue(SessionQueueBase):
             )
             self.__conn.commit()
             if current_queue_item is not None and current_queue_item.batch_id in batch_ids:
-                self.__invoker.services.events.emit_session_canceled(
-                    queue_item_id=current_queue_item.item_id,
-                    queue_id=current_queue_item.queue_id,
-                    queue_batch_id=current_queue_item.batch_id,
-                    graph_execution_state_id=current_queue_item.session_id,
-                )
                 batch_status = self.get_batch_status(queue_id=queue_id, batch_id=current_queue_item.batch_id)
                 queue_status = self.get_queue_status(queue_id=queue_id)
                 self.__invoker.services.events.emit_queue_item_status_changed(
-                    session_queue_item=current_queue_item,
-                    batch_status=batch_status,
-                    queue_status=queue_status,
+                    current_queue_item, batch_status, queue_status
                 )
         except Exception:
             self.__conn.rollback()
@@ -521,18 +460,10 @@ class SqliteSessionQueue(SessionQueueBase):
             )
             self.__conn.commit()
             if current_queue_item is not None and current_queue_item.queue_id == queue_id:
-                self.__invoker.services.events.emit_session_canceled(
-                    queue_item_id=current_queue_item.item_id,
-                    queue_id=current_queue_item.queue_id,
-                    queue_batch_id=current_queue_item.batch_id,
-                    graph_execution_state_id=current_queue_item.session_id,
-                )
                 batch_status = self.get_batch_status(queue_id=queue_id, batch_id=current_queue_item.batch_id)
                 queue_status = self.get_queue_status(queue_id=queue_id)
                 self.__invoker.services.events.emit_queue_item_status_changed(
-                    session_queue_item=current_queue_item,
-                    batch_status=batch_status,
-                    queue_status=queue_status,
+                    current_queue_item, batch_status, queue_status
                 )
         except Exception:
             self.__conn.rollback()
@@ -562,6 +493,29 @@ class SqliteSessionQueue(SessionQueueBase):
             raise SessionQueueItemNotFoundError(f"No queue item with id {item_id}")
         return SessionQueueItem.queue_item_from_dict(dict(result))
 
+    def set_queue_item_session(self, item_id: int, session: GraphExecutionState) -> SessionQueueItem:
+        try:
+            # Use exclude_none so we don't end up with a bunch of nulls in the graph - this can cause validation errors
+            # when the graph is loaded. Graph execution occurs purely in memory - the session saved here is not referenced
+            # during execution.
+            session_json = session.model_dump_json(warnings=False, exclude_none=True)
+            self.__lock.acquire()
+            self.__cursor.execute(
+                """--sql
+                UPDATE session_queue
+                SET session = ?
+                WHERE item_id = ?
+                """,
+                (session_json, item_id),
+            )
+            self.__conn.commit()
+        except Exception:
+            self.__conn.rollback()
+            raise
+        finally:
+            self.__lock.release()
+        return self.get_queue_item(item_id)
+
     def list_queue_items(
         self,
         queue_id: str,
@@ -578,7 +532,9 @@ class SqliteSessionQueue(SessionQueueBase):
                     status,
                     priority,
                     field_values,
-                    error,
+                    error_type,
+                    error_message,
+                    error_traceback,
                     created_at,
                     updated_at,
                     completed_at,

invokeai/app/services/shared/graph.py

@@ -2,17 +2,19 @@
 
 import copy
 import itertools
-from typing import Annotated, Any, Optional, TypeVar, Union, get_args, get_origin, get_type_hints
+from typing import Any, Optional, TypeVar, Union, get_args, get_origin, get_type_hints
 
 import networkx as nx
 from pydantic import (
     BaseModel,
+    GetCoreSchemaHandler,
     GetJsonSchemaHandler,
+    ValidationError,
     field_validator,
 )
 from pydantic.fields import Field
 from pydantic.json_schema import JsonSchemaValue
-from pydantic_core import CoreSchema
+from pydantic_core import core_schema
 
 # Importing * is bad karma but needed here for node detection
 from invokeai.app.invocations import *  # noqa: F401 F403
@@ -190,6 +192,39 @@ class UnknownGraphValidationError(ValueError):
     pass
 
 
+class NodeInputError(ValueError):
+    """Raised when a node fails preparation. This occurs when a node's inputs are being set from its incomers, but an
+    input fails validation.
+
+    Attributes:
+        node: The node that failed preparation. Note: only successfully set fields will be accurate. Review the error to
+            determine which field caused the failure.
+    """
+
+    def __init__(self, node: BaseInvocation, e: ValidationError):
+        self.original_error = e
+        self.node = node
+        # When preparing a node, we set each input one-at-a-time. We may thus safely assume that the first error
+        # represents the first input that failed.
+        self.failed_input = loc_to_dot_sep(e.errors()[0]["loc"])
+        super().__init__(f"Node {node.id} has invalid incoming input for {self.failed_input}")
+
+
+def loc_to_dot_sep(loc: tuple[Union[str, int], ...]) -> str:
+    """Helper to pretty-print pydantic error locations as dot-separated strings.
+    Taken from https://docs.pydantic.dev/latest/errors/errors/#customize-error-messages
+    """
+    path = ""
+    for i, x in enumerate(loc):
+        if isinstance(x, str):
+            if i > 0:
+                path += "."
+            path += x
+        else:
+            path += f"[{x}]"
+    return path
+
+
 @invocation_output("iterate_output")
 class IterateInvocationOutput(BaseInvocationOutput):
     """Used to connect iteration outputs. Will be expanded to a specific output."""
@@ -243,73 +278,58 @@ class CollectInvocation(BaseInvocation):
         return CollectInvocationOutput(collection=copy.copy(self.collection))
 
 
+class AnyInvocation(BaseInvocation):
+    @classmethod
+    def __get_pydantic_core_schema__(cls, source_type: Any, handler: GetCoreSchemaHandler) -> core_schema.CoreSchema:
+        def validate_invocation(v: Any) -> "AnyInvocation":
+            return BaseInvocation.get_typeadapter().validate_python(v)
+
+        return core_schema.no_info_plain_validator_function(validate_invocation)
+
+    @classmethod
+    def __get_pydantic_json_schema__(
+        cls, core_schema: core_schema.CoreSchema, handler: GetJsonSchemaHandler
+    ) -> JsonSchemaValue:
+        # Nodes are too powerful, we have to make our own OpenAPI schema manually
+        # No but really, because the schema is dynamic depending on loaded nodes, we need to generate it manually
+        oneOf: list[dict[str, str]] = []
+        names = [i.__name__ for i in BaseInvocation.get_invocations()]
+        for name in sorted(names):
+            oneOf.append({"$ref": f"#/components/schemas/{name}"})
+        return {"oneOf": oneOf}
+
+
+class AnyInvocationOutput(BaseInvocationOutput):
+    @classmethod
+    def __get_pydantic_core_schema__(cls, source_type: Any, handler: GetCoreSchemaHandler):
+        def validate_invocation_output(v: Any) -> "AnyInvocationOutput":
+            return BaseInvocationOutput.get_typeadapter().validate_python(v)
+
+        return core_schema.no_info_plain_validator_function(validate_invocation_output)
+
+    @classmethod
+    def __get_pydantic_json_schema__(
+        cls, core_schema: core_schema.CoreSchema, handler: GetJsonSchemaHandler
+    ) -> JsonSchemaValue:
+        # Nodes are too powerful, we have to make our own OpenAPI schema manually
+        # No but really, because the schema is dynamic depending on loaded nodes, we need to generate it manually
+
+        oneOf: list[dict[str, str]] = []
+        names = [i.__name__ for i in BaseInvocationOutput.get_outputs()]
+        for name in sorted(names):
+            oneOf.append({"$ref": f"#/components/schemas/{name}"})
+        return {"oneOf": oneOf}
+
+
 class Graph(BaseModel):
     id: str = Field(description="The id of this graph", default_factory=uuid_string)
     # TODO: use a list (and never use dict in a BaseModel) because pydantic/fastapi hates me
-    nodes: dict[str, BaseInvocation] = Field(description="The nodes in this graph", default_factory=dict)
+    nodes: dict[str, AnyInvocation] = Field(description="The nodes in this graph", default_factory=dict)
     edges: list[Edge] = Field(
         description="The connections between nodes and their fields in this graph",
         default_factory=list,
     )
 
-    @field_validator("nodes", mode="plain")
-    @classmethod
-    def validate_nodes(cls, v: dict[str, Any]):
-        """Validates the nodes in the graph by retrieving a union of all node types and validating each node."""
-
-        # Invocations register themselves as their python modules are executed. The union of all invocations is
-        # constructed at runtime. We use pydantic to validate `Graph.nodes` using that union.
-        #
-        # It's possible that when `graph.py` is executed, not all invocation-containing modules will have executed. If
-        # we construct the invocation union as `graph.py` is executed, we may miss some invocations. Those missing
-        # invocations will cause a graph to fail if they are used.
-        #
-        # We can get around this by validating the nodes in the graph using a "plain" validator, which overrides the
-        # pydantic validation entirely. This allows us to validate the nodes using the union of invocations at runtime.
-        #
-        # This same pattern is used in `GraphExecutionState`.
-
-        nodes: dict[str, BaseInvocation] = {}
-        typeadapter = BaseInvocation.get_typeadapter()
-        for node_id, node in v.items():
-            nodes[node_id] = typeadapter.validate_python(node)
-        return nodes
-
-    @classmethod
-    def __get_pydantic_json_schema__(cls, core_schema: CoreSchema, handler: GetJsonSchemaHandler) -> JsonSchemaValue:
-        # We use a "plain" validator to validate the nodes in the graph. Pydantic is unable to create a JSON Schema for
-        # fields that use "plain" validators, so we have to hack around this. Also, we need to add all invocations to
-        # the generated schema as options for the `nodes` field.
-        #
-        # The workaround is to create a new BaseModel that has the same fields as `Graph` but without the validator and
-        # with the invocation union as the type for the `nodes` field. Pydantic then generates the JSON Schema as
-        # expected.
-        #
-        # You might be tempted to do something like this:
-        #
-        # ```py
-        # cloned_model = create_model(cls.__name__, __base__=cls, nodes=...)
-        # delattr(cloned_model, "validate_nodes")
-        # cloned_model.model_rebuild(force=True)
-        # json_schema = handler(cloned_model.__pydantic_core_schema__)
-        # ```
-        #
-        # Unfortunately, this does not work. Calling `handler` here results in infinite recursion as pydantic attempts
-        # to build the JSON Schema for the cloned model. Instead, we have to manually clone the model.
-        #
-        # This same pattern is used in `GraphExecutionState`.
-
-        class Graph(BaseModel):
-            id: Optional[str] = Field(default=None, description="The id of this graph")
-            nodes: dict[
-                str, Annotated[Union[tuple(BaseInvocation._invocation_classes)], Field(discriminator="type")]
-            ] = Field(description="The nodes in this graph")
-            edges: list[Edge] = Field(description="The connections between nodes and their fields in this graph")
-
-        json_schema = handler(Graph.__pydantic_core_schema__)
-        json_schema = handler.resolve_ref_schema(json_schema)
-        return json_schema
-
     def add_node(self, node: BaseInvocation) -> None:
         """Adds a node to a graph
 
@@ -632,7 +652,7 @@ class Graph(BaseModel):
         output_fields = [get_input_field(self.get_node(e.node_id), e.field) for e in outputs]
 
         # Input type must be a list
-        if get_origin(input_field) != list:
+        if get_origin(input_field) is not list:
             return False
 
         # Validate that all outputs match the input type
@@ -740,7 +760,7 @@ class GraphExecutionState(BaseModel):
     )
 
     # The results of executed nodes
-    results: dict[str, BaseInvocationOutput] = Field(description="The results of node executions", default_factory=dict)
+    results: dict[str, AnyInvocationOutput] = Field(description="The results of node executions", default_factory=dict)
 
     # Errors raised when executing nodes
     errors: dict[str, str] = Field(description="Errors raised when executing nodes", default_factory=dict)
@@ -757,52 +777,12 @@ class GraphExecutionState(BaseModel):
         default_factory=dict,
     )
 
-    @field_validator("results", mode="plain")
-    @classmethod
-    def validate_results(cls, v: dict[str, BaseInvocationOutput]):
-        """Validates the results in the GES by retrieving a union of all output types and validating each result."""
-
-        # See the comment in `Graph.validate_nodes` for an explanation of this logic.
-        results: dict[str, BaseInvocationOutput] = {}
-        typeadapter = BaseInvocationOutput.get_typeadapter()
-        for result_id, result in v.items():
-            results[result_id] = typeadapter.validate_python(result)
-        return results
-
     @field_validator("graph")
     def graph_is_valid(cls, v: Graph):
         """Validates that the graph is valid"""
         v.validate_self()
         return v
 
-    @classmethod
-    def __get_pydantic_json_schema__(cls, core_schema: CoreSchema, handler: GetJsonSchemaHandler) -> JsonSchemaValue:
-        # See the comment in `Graph.__get_pydantic_json_schema__` for an explanation of this logic.
-        class GraphExecutionState(BaseModel):
-            """Tracks the state of a graph execution"""
-
-            id: str = Field(description="The id of the execution state")
-            graph: Graph = Field(description="The graph being executed")
-            execution_graph: Graph = Field(description="The expanded graph of activated and executed nodes")
-            executed: set[str] = Field(description="The set of node ids that have been executed")
-            executed_history: list[str] = Field(
-                description="The list of node ids that have been executed, in order of execution"
-            )
-            results: dict[
-                str, Annotated[Union[tuple(BaseInvocationOutput._output_classes)], Field(discriminator="type")]
-            ] = Field(description="The results of node executions")
-            errors: dict[str, str] = Field(description="Errors raised when executing nodes")
-            prepared_source_mapping: dict[str, str] = Field(
-                description="The map of prepared nodes to original graph nodes"
-            )
-            source_prepared_mapping: dict[str, set[str]] = Field(
-                description="The map of original graph nodes to prepared nodes"
-            )
-
-        json_schema = handler(GraphExecutionState.__pydantic_core_schema__)
-        json_schema = handler.resolve_ref_schema(json_schema)
-        return json_schema
-
     def next(self) -> Optional[BaseInvocation]:
         """Gets the next node ready to execute."""
 
@@ -821,7 +801,10 @@ class GraphExecutionState(BaseModel):
 
         # Get values from edges
         if next_node is not None:
-            self._prepare_inputs(next_node)
+            try:
+                self._prepare_inputs(next_node)
+            except ValidationError as e:
+                raise NodeInputError(next_node, e)
 
         # If next is still none, there's no next node, return None
         return next_node

invokeai/app/services/shared/invocation_context.py

@@ -1,9 +1,10 @@
-import threading
 from dataclasses import dataclass
 from pathlib import Path
-from typing import TYPE_CHECKING, Optional, Union
+from typing import TYPE_CHECKING, Callable, Optional, Union
 
+import torch
 from PIL.Image import Image
+from pydantic.networks import AnyHttpUrl
 from torch import Tensor
 
 from invokeai.app.invocations.constants import IMAGE_MODES
@@ -15,15 +16,24 @@ from invokeai.app.services.images.images_common import ImageDTO
 from invokeai.app.services.invocation_services import InvocationServices
 from invokeai.app.services.model_records.model_records_base import UnknownModelException
 from invokeai.app.util.step_callback import stable_diffusion_step_callback
-from invokeai.backend.model_manager.config import AnyModelConfig, BaseModelType, ModelFormat, ModelType, SubModelType
-from invokeai.backend.model_manager.load.load_base import LoadedModel
+from invokeai.backend.model_manager.config import (
+    AnyModel,
+    AnyModelConfig,
+    BaseModelType,
+    ModelFormat,
+    ModelType,
+    SubModelType,
+)
+from invokeai.backend.model_manager.load.load_base import LoadedModel, LoadedModelWithoutConfig
 from invokeai.backend.stable_diffusion.diffusers_pipeline import PipelineIntermediateState
 from invokeai.backend.stable_diffusion.diffusion.conditioning_data import ConditioningFieldData
+from invokeai.backend.util.devices import TorchDevice
 
 if TYPE_CHECKING:
     from invokeai.app.invocations.baseinvocation import BaseInvocation
     from invokeai.app.invocations.model import ModelIdentifierField
     from invokeai.app.services.session_queue.session_queue_common import SessionQueueItem
+    from invokeai.backend.model_manager.load.model_cache.model_cache_base import ModelCacheBase
 
 """
 The InvocationContext provides access to various services and data about the current invocation.
@@ -180,9 +190,9 @@ class ImagesInterface(InvocationContextInterface):
         # If `metadata` is provided directly, use that. Else, use the metadata provided by `WithMetadata`, falling back to None.
         metadata_ = None
         if metadata:
-            metadata_ = metadata
-        elif isinstance(self._data.invocation, WithMetadata):
-            metadata_ = self._data.invocation.metadata
+            metadata_ = metadata.model_dump_json()
+        elif isinstance(self._data.invocation, WithMetadata) and self._data.invocation.metadata:
+            metadata_ = self._data.invocation.metadata.model_dump_json()
 
         # If `board_id` is provided directly, use that. Else, use the board provided by `WithBoard`, falling back to None.
         board_id_ = None
@@ -191,6 +201,14 @@ class ImagesInterface(InvocationContextInterface):
         elif isinstance(self._data.invocation, WithBoard) and self._data.invocation.board:
             board_id_ = self._data.invocation.board.board_id
 
+        workflow_ = None
+        if self._data.queue_item.workflow:
+            workflow_ = self._data.queue_item.workflow.model_dump_json()
+
+        graph_ = None
+        if self._data.queue_item.session.graph:
+            graph_ = self._data.queue_item.session.graph.model_dump_json()
+
         return self._services.images.create(
             image=image,
             is_intermediate=self._data.invocation.is_intermediate,
@@ -198,7 +216,8 @@ class ImagesInterface(InvocationContextInterface):
             board_id=board_id_,
             metadata=metadata_,
             image_origin=ResourceOrigin.INTERNAL,
-            workflow=self._data.queue_item.workflow,
+            workflow=workflow_,
+            graph=graph_,
             session_id=self._data.queue_item.session_id,
             node_id=self._data.invocation.id,
         )
@@ -307,13 +326,14 @@ class ConditioningInterface(InvocationContextInterface):
         Returns:
             The loaded conditioning data.
         """
-
         return self._services.conditioning.load(name)
 
 
 class ModelsInterface(InvocationContextInterface):
+    """Common API for loading, downloading and managing models."""
+
     def exists(self, identifier: Union[str, "ModelIdentifierField"]) -> bool:
-        """Checks if a model exists.
+        """Check if a model exists.
 
         Args:
             identifier: The key or ModelField representing the model.
@@ -323,13 +343,13 @@ class ModelsInterface(InvocationContextInterface):
         """
         if isinstance(identifier, str):
             return self._services.model_manager.store.exists(identifier)
-
-        return self._services.model_manager.store.exists(identifier.key)
+        else:
+            return self._services.model_manager.store.exists(identifier.key)
 
     def load(
         self, identifier: Union[str, "ModelIdentifierField"], submodel_type: Optional[SubModelType] = None
     ) -> LoadedModel:
-        """Loads a model.
+        """Load a model.
 
         Args:
             identifier: The key or ModelField representing the model.
@@ -344,16 +364,16 @@ class ModelsInterface(InvocationContextInterface):
 
         if isinstance(identifier, str):
             model = self._services.model_manager.store.get_model(identifier)
-            return self._services.model_manager.load.load_model(model, submodel_type, self._data)
+            return self._services.model_manager.load.load_model(model, submodel_type)
         else:
             _submodel_type = submodel_type or identifier.submodel_type
             model = self._services.model_manager.store.get_model(identifier.key)
-            return self._services.model_manager.load.load_model(model, _submodel_type, self._data)
+            return self._services.model_manager.load.load_model(model, _submodel_type)
 
     def load_by_attrs(
         self, name: str, base: BaseModelType, type: ModelType, submodel_type: Optional[SubModelType] = None
     ) -> LoadedModel:
-        """Loads a model by its attributes.
+        """Load a model by its attributes.
 
         Args:
             name: Name of the model.
@@ -373,10 +393,10 @@ class ModelsInterface(InvocationContextInterface):
         if len(configs) > 1:
             raise ValueError(f"More than one model found with name {name}, base {base}, and type {type}")
 
-        return self._services.model_manager.load.load_model(configs[0], submodel_type, self._data)
+        return self._services.model_manager.load.load_model(configs[0], submodel_type)
 
     def get_config(self, identifier: Union[str, "ModelIdentifierField"]) -> AnyModelConfig:
-        """Gets a model's config.
+        """Get a model's config.
 
         Args:
             identifier: The key or ModelField representing the model.
@@ -386,11 +406,11 @@ class ModelsInterface(InvocationContextInterface):
         """
         if isinstance(identifier, str):
             return self._services.model_manager.store.get_model(identifier)
-
-        return self._services.model_manager.store.get_model(identifier.key)
+        else:
+            return self._services.model_manager.store.get_model(identifier.key)
 
     def search_by_path(self, path: Path) -> list[AnyModelConfig]:
-        """Searches for models by path.
+        """Search for models by path.
 
         Args:
             path: The path to search for.
@@ -407,7 +427,7 @@ class ModelsInterface(InvocationContextInterface):
         type: Optional[ModelType] = None,
         format: Optional[ModelFormat] = None,
     ) -> list[AnyModelConfig]:
-        """Searches for models by attributes.
+        """Search for models by attributes.
 
         Args:
             name: The name to search for (exact match).
@@ -426,6 +446,72 @@ class ModelsInterface(InvocationContextInterface):
             model_format=format,
         )
 
+    def download_and_cache_model(
+        self,
+        source: str | AnyHttpUrl,
+    ) -> Path:
+        """
+        Download the model file located at source to the models cache and return its Path.
+
+        This can be used to single-file install models and other resources of arbitrary types
+        which should not get registered with the database. If the model is already
+        installed, the cached path will be returned. Otherwise it will be downloaded.
+
+        Args:
+            source: A URL that points to the model, or a huggingface repo_id.
+
+        Returns:
+            Path to the downloaded model
+        """
+        return self._services.model_manager.install.download_and_cache_model(source=source)
+
+    def load_local_model(
+        self,
+        model_path: Path,
+        loader: Optional[Callable[[Path], AnyModel]] = None,
+    ) -> LoadedModelWithoutConfig:
+        """
+        Load the model file located at the indicated path
+
+        If a loader callable is provided, it will be invoked to load the model. Otherwise,
+        `safetensors.torch.load_file()` or `torch.load()` will be called to load the model.
+
+        Be aware that the LoadedModelWithoutConfig object has no `config` attribute
+
+        Args:
+            path: A model Path
+            loader: A Callable that expects a Path and returns a dict[str|int, Any]
+
+        Returns:
+            A LoadedModelWithoutConfig object.
+        """
+        return self._services.model_manager.load.load_model_from_path(model_path=model_path, loader=loader)
+
+    def load_remote_model(
+        self,
+        source: str | AnyHttpUrl,
+        loader: Optional[Callable[[Path], AnyModel]] = None,
+    ) -> LoadedModelWithoutConfig:
+        """
+        Download, cache, and load the model file located at the indicated URL or repo_id.
+
+        If the model is already downloaded, it will be loaded from the cache.
+
+        If the a loader callable is provided, it will be invoked to load the model. Otherwise,
+        `safetensors.torch.load_file()` or `torch.load()` will be called to load the model.
+
+        Be aware that the LoadedModelWithoutConfig object has no `config` attribute
+
+        Args:
+            source: A URL or huggingface repoid.
+            loader: A Callable that expects a Path and returns a dict[str|int, Any]
+
+        Returns:
+            A LoadedModelWithoutConfig object.
+        """
+        model_path = self._services.model_manager.install.download_and_cache_model(source=str(source))
+        return self._services.model_manager.load.load_model_from_path(model_path=model_path, loader=loader)
+
 
 class ConfigInterface(InvocationContextInterface):
     def get(self) -> InvokeAIAppConfig:
@@ -440,10 +526,10 @@ class ConfigInterface(InvocationContextInterface):
 
 class UtilInterface(InvocationContextInterface):
     def __init__(
-        self, services: InvocationServices, data: InvocationContextData, cancel_event: threading.Event
+        self, services: InvocationServices, data: InvocationContextData, is_canceled: Callable[[], bool]
     ) -> None:
         super().__init__(services, data)
-        self._cancel_event = cancel_event
+        self._is_canceled = is_canceled
 
     def is_canceled(self) -> bool:
         """Checks if the current session has been canceled.
@@ -451,7 +537,7 @@ class UtilInterface(InvocationContextInterface):
         Returns:
             True if the current session has been canceled, False if not.
         """
-        return self._cancel_event.is_set()
+        return self._is_canceled()
 
     def sd_step_callback(self, intermediate_state: PipelineIntermediateState, base_model: BaseModelType) -> None:
         """
@@ -473,6 +559,28 @@ class UtilInterface(InvocationContextInterface):
             is_canceled=self.is_canceled,
         )
 
+    def torch_device(self) -> torch.device:
+        """
+        Return a torch device to use in the current invocation.
+
+        Returns:
+            A torch.device not currently in use by the system.
+        """
+        ram_cache: "ModelCacheBase[AnyModel]" = self._services.model_manager.load.ram_cache
+        return ram_cache.get_execution_device()
+
+    def torch_dtype(self, device: Optional[torch.device] = None) -> torch.dtype:
+        """
+        Return a precision type to use with the current invocation and torch device.
+
+        Args:
+            device: Optional device.
+
+        Returns:
+            A torch.dtype suited for the current device.
+        """
+        return TorchDevice.choose_torch_dtype(device)
+
 
 class InvocationContext:
     """Provides access to various services and data for the current invocation.
@@ -526,7 +634,7 @@ class InvocationContext:
 def build_invocation_context(
     services: InvocationServices,
     data: InvocationContextData,
-    cancel_event: threading.Event,
+    is_canceled: Callable[[], bool],
 ) -> InvocationContext:
     """Builds the invocation context for a specific invocation execution.
 
@@ -543,7 +651,7 @@ def build_invocation_context(
     tensors = TensorsInterface(services=services, data=data)
     models = ModelsInterface(services=services, data=data)
     config = ConfigInterface(services=services, data=data)
-    util = UtilInterface(services=services, data=data, cancel_event=cancel_event)
+    util = UtilInterface(services=services, data=data, is_canceled=is_canceled)
     conditioning = ConditioningInterface(services=services, data=data)
     boards = BoardsInterface(services=services, data=data)
 

invokeai/app/services/shared/sqlite/sqlite_util.py

@@ -12,6 +12,8 @@ from invokeai.app.services.shared.sqlite_migrator.migrations.migration_6 import
 from invokeai.app.services.shared.sqlite_migrator.migrations.migration_7 import build_migration_7
 from invokeai.app.services.shared.sqlite_migrator.migrations.migration_8 import build_migration_8
 from invokeai.app.services.shared.sqlite_migrator.migrations.migration_9 import build_migration_9
+from invokeai.app.services.shared.sqlite_migrator.migrations.migration_10 import build_migration_10
+from invokeai.app.services.shared.sqlite_migrator.migrations.migration_11 import build_migration_11
 from invokeai.app.services.shared.sqlite_migrator.sqlite_migrator_impl import SqliteMigrator
 
 
@@ -41,6 +43,8 @@ def init_db(config: InvokeAIAppConfig, logger: Logger, image_files: ImageFileSto
     migrator.register_migration(build_migration_7())
     migrator.register_migration(build_migration_8(app_config=config))
     migrator.register_migration(build_migration_9())
+    migrator.register_migration(build_migration_10())
+    migrator.register_migration(build_migration_11(app_config=config, logger=logger))
     migrator.run_migrations()
 
     return db

invokeai/app/services/shared/sqlite_migrator/migrations/migration_10.py

@@ -0,0 +1,35 @@
+import sqlite3
+
+from invokeai.app.services.shared.sqlite_migrator.sqlite_migrator_common import Migration
+
+
+class Migration10Callback:
+    def __call__(self, cursor: sqlite3.Cursor) -> None:
+        self._update_error_cols(cursor)
+
+    def _update_error_cols(self, cursor: sqlite3.Cursor) -> None:
+        """
+        - Adds `error_type` and `error_message` columns to the session queue table.
+        - Renames the `error` column to `error_traceback`.
+        """
+
+        cursor.execute("ALTER TABLE session_queue ADD COLUMN error_type TEXT;")
+        cursor.execute("ALTER TABLE session_queue ADD COLUMN error_message TEXT;")
+        cursor.execute("ALTER TABLE session_queue RENAME COLUMN error TO error_traceback;")
+
+
+def build_migration_10() -> Migration:
+    """
+    Build the migration from database version 9 to 10.
+
+    This migration does the following:
+    - Adds `error_type` and `error_message` columns to the session queue table.
+    - Renames the `error` column to `error_traceback`.
+    """
+    migration_10 = Migration(
+        from_version=9,
+        to_version=10,
+        callback=Migration10Callback(),
+    )
+
+    return migration_10

invokeai/app/services/shared/sqlite_migrator/migrations/migration_11.py

@@ -0,0 +1,75 @@
+import shutil
+import sqlite3
+from logging import Logger
+
+from invokeai.app.services.config import InvokeAIAppConfig
+from invokeai.app.services.shared.sqlite_migrator.sqlite_migrator_common import Migration
+
+LEGACY_CORE_MODELS = [
+    # OpenPose
+    "any/annotators/dwpose/yolox_l.onnx",
+    "any/annotators/dwpose/dw-ll_ucoco_384.onnx",
+    # DepthAnything
+    "any/annotators/depth_anything/depth_anything_vitl14.pth",
+    "any/annotators/depth_anything/depth_anything_vitb14.pth",
+    "any/annotators/depth_anything/depth_anything_vits14.pth",
+    # Lama inpaint
+    "core/misc/lama/lama.pt",
+    # RealESRGAN upscale
+    "core/upscaling/realesrgan/RealESRGAN_x4plus.pth",
+    "core/upscaling/realesrgan/RealESRGAN_x4plus_anime_6B.pth",
+    "core/upscaling/realesrgan/ESRGAN_SRx4_DF2KOST_official-ff704c30.pth",
+    "core/upscaling/realesrgan/RealESRGAN_x2plus.pth",
+]
+
+
+class Migration11Callback:
+    def __init__(self, app_config: InvokeAIAppConfig, logger: Logger) -> None:
+        self._app_config = app_config
+        self._logger = logger
+
+    def __call__(self, cursor: sqlite3.Cursor) -> None:
+        self._remove_convert_cache()
+        self._remove_downloaded_models()
+        self._remove_unused_core_models()
+
+    def _remove_convert_cache(self) -> None:
+        """Rename models/.cache to models/.convert_cache."""
+        self._logger.info("Removing .cache directory. Converted models will now be cached in .convert_cache.")
+        legacy_convert_path = self._app_config.root_path / "models" / ".cache"
+        shutil.rmtree(legacy_convert_path, ignore_errors=True)
+
+    def _remove_downloaded_models(self) -> None:
+        """Remove models from their old locations; they will re-download when needed."""
+        self._logger.info(
+            "Removing legacy just-in-time models. Downloaded models will now be cached in .download_cache."
+        )
+        for model_path in LEGACY_CORE_MODELS:
+            legacy_dest_path = self._app_config.models_path / model_path
+            legacy_dest_path.unlink(missing_ok=True)
+
+    def _remove_unused_core_models(self) -> None:
+        """Remove unused core models and their directories."""
+        self._logger.info("Removing defunct core models.")
+        for dir in ["face_restoration", "misc", "upscaling"]:
+            path_to_remove = self._app_config.models_path / "core" / dir
+            shutil.rmtree(path_to_remove, ignore_errors=True)
+        shutil.rmtree(self._app_config.models_path / "any" / "annotators", ignore_errors=True)
+
+
+def build_migration_11(app_config: InvokeAIAppConfig, logger: Logger) -> Migration:
+    """
+    Build the migration from database version 10 to 11.
+
+    This migration does the following:
+    - Moves "core" models previously downloaded with download_with_progress_bar() into new
+      "models/.download_cache" directory.
+    - Renames "models/.cache" to "models/.convert_cache".
+    """
+    migration_11 = Migration(
+        from_version=10,
+        to_version=11,
+        callback=Migration11Callback(app_config=app_config, logger=logger),
+    )
+
+    return migration_11

invokeai/app/util/custom_openapi.py

@@ -0,0 +1,116 @@
+from typing import Any, Callable, Optional
+
+from fastapi import FastAPI
+from fastapi.openapi.utils import get_openapi
+from pydantic.json_schema import models_json_schema
+
+from invokeai.app.invocations.baseinvocation import BaseInvocation, BaseInvocationOutput, UIConfigBase
+from invokeai.app.invocations.fields import InputFieldJSONSchemaExtra, OutputFieldJSONSchemaExtra
+from invokeai.app.invocations.model import ModelIdentifierField
+from invokeai.app.services.events.events_common import EventBase
+from invokeai.app.services.session_processor.session_processor_common import ProgressImage
+
+
+def move_defs_to_top_level(openapi_schema: dict[str, Any], component_schema: dict[str, Any]) -> None:
+    """Moves a component schema's $defs to the top level of the openapi schema. Useful when generating a schema
+    for a single model that needs to be added back to the top level of the schema. Mutates openapi_schema and
+    component_schema."""
+
+    defs = component_schema.pop("$defs", {})
+    for schema_key, json_schema in defs.items():
+        if schema_key in openapi_schema["components"]["schemas"]:
+            continue
+        openapi_schema["components"]["schemas"][schema_key] = json_schema
+
+
+def get_openapi_func(
+    app: FastAPI, post_transform: Optional[Callable[[dict[str, Any]], dict[str, Any]]] = None
+) -> Callable[[], dict[str, Any]]:
+    """Gets the OpenAPI schema generator function.
+
+    Args:
+        app (FastAPI): The FastAPI app to generate the schema for.
+        post_transform (Optional[Callable[[dict[str, Any]], dict[str, Any]]], optional): A function to apply to the
+            generated schema before returning it. Defaults to None.
+
+    Returns:
+        Callable[[], dict[str, Any]]: The OpenAPI schema generator function. When first called, the generated schema is
+            cached in `app.openapi_schema`. On subsequent calls, the cached schema is returned. This caching behaviour
+            matches FastAPI's default schema generation caching.
+    """
+
+    def openapi() -> dict[str, Any]:
+        if app.openapi_schema:
+            return app.openapi_schema
+
+        openapi_schema = get_openapi(
+            title=app.title,
+            description="An API for invoking AI image operations",
+            version="1.0.0",
+            routes=app.routes,
+            separate_input_output_schemas=False,  # https://fastapi.tiangolo.com/how-to/separate-openapi-schemas/
+        )
+
+        # We'll create a map of invocation type to output schema to make some types simpler on the client.
+        invocation_output_map_properties: dict[str, Any] = {}
+        invocation_output_map_required: list[str] = []
+
+        # We need to manually add all outputs to the schema - pydantic doesn't add them because they aren't used directly.
+        for output in BaseInvocationOutput.get_outputs():
+            json_schema = output.model_json_schema(mode="serialization", ref_template="#/components/schemas/{model}")
+            move_defs_to_top_level(openapi_schema, json_schema)
+            openapi_schema["components"]["schemas"][output.__name__] = json_schema
+
+        # Technically, invocations are added to the schema by pydantic, but we still need to manually set their output
+        # property, so we'll just do it all manually.
+        for invocation in BaseInvocation.get_invocations():
+            json_schema = invocation.model_json_schema(
+                mode="serialization", ref_template="#/components/schemas/{model}"
+            )
+            move_defs_to_top_level(openapi_schema, json_schema)
+            output_title = invocation.get_output_annotation().__name__
+            outputs_ref = {"$ref": f"#/components/schemas/{output_title}"}
+            json_schema["output"] = outputs_ref
+            openapi_schema["components"]["schemas"][invocation.__name__] = json_schema
+
+            # Add this invocation and its output to the output map
+            invocation_type = invocation.get_type()
+            invocation_output_map_properties[invocation_type] = json_schema["output"]
+            invocation_output_map_required.append(invocation_type)
+
+        # Add the output map to the schema
+        openapi_schema["components"]["schemas"]["InvocationOutputMap"] = {
+            "type": "object",
+            "properties": invocation_output_map_properties,
+            "required": invocation_output_map_required,
+        }
+
+        # Some models don't end up in the schemas as standalone definitions because they aren't used directly in the API.
+        # We need to add them manually here. WARNING: Pydantic can choke if you call `model.model_json_schema()` to get
+        # a schema. This has something to do with schema refs - not totally clear. For whatever reason, using
+        # `models_json_schema` seems to work fine.
+        additional_models = [
+            *EventBase.get_events(),
+            UIConfigBase,
+            InputFieldJSONSchemaExtra,
+            OutputFieldJSONSchemaExtra,
+            ModelIdentifierField,
+            ProgressImage,
+        ]
+
+        additional_schemas = models_json_schema(
+            [(m, "serialization") for m in additional_models],
+            ref_template="#/components/schemas/{model}",
+        )
+        # additional_schemas[1] is a dict of $defs that we need to add to the top level of the schema
+        move_defs_to_top_level(openapi_schema, additional_schemas[1])
+
+        if post_transform is not None:
+            openapi_schema = post_transform(openapi_schema)
+
+        openapi_schema["components"]["schemas"] = dict(sorted(openapi_schema["components"]["schemas"].items()))
+
+        app.openapi_schema = openapi_schema
+        return app.openapi_schema
+
+    return openapi

invokeai/app/util/download_with_progress.py

@@ -1,51 +0,0 @@
-from pathlib import Path
-from urllib import request
-
-from tqdm import tqdm
-
-from invokeai.backend.util.logging import InvokeAILogger
-
-
-class ProgressBar:
-    """Simple progress bar for urllib.request.urlretrieve using tqdm."""
-
-    def __init__(self, model_name: str = "file"):
-        self.pbar = None
-        self.name = model_name
-
-    def __call__(self, block_num: int, block_size: int, total_size: int):
-        if not self.pbar:
-            self.pbar = tqdm(
-                desc=self.name,
-                initial=0,
-                unit="iB",
-                unit_scale=True,
-                unit_divisor=1000,
-                total=total_size,
-            )
-        self.pbar.update(block_size)
-
-
-def download_with_progress_bar(name: str, url: str, dest_path: Path) -> bool:
-    """Download a file from a URL to a destination path, with a progress bar.
-    If the file already exists, it will not be downloaded again.
-
-    Exceptions are not caught.
-
-    Args:
-        name (str): Name of the file being downloaded.
-        url (str): URL to download the file from.
-        dest_path (Path): Destination path to save the file to.
-
-    Returns:
-        bool: True if the file was downloaded, False if it already existed.
-    """
-    if dest_path.exists():
-        return False  # already downloaded
-
-    InvokeAILogger.get_logger().info(f"Downloading {name}...")
-
-    dest_path.parent.mkdir(parents=True, exist_ok=True)
-    request.urlretrieve(url, dest_path, ProgressBar(name))
-
-    return True

invokeai/app/util/step_callback.py

@@ -1,4 +1,4 @@
-from typing import TYPE_CHECKING, Callable
+from typing import TYPE_CHECKING, Callable, Optional
 
 import torch
 from PIL import Image
@@ -13,8 +13,36 @@ if TYPE_CHECKING:
     from invokeai.app.services.events.events_base import EventServiceBase
     from invokeai.app.services.shared.invocation_context import InvocationContextData
 
+# fast latents preview matrix for sdxl
+# generated by @StAlKeR7779
+SDXL_LATENT_RGB_FACTORS = [
+    #   R        G        B
+    [0.3816, 0.4930, 0.5320],
+    [-0.3753, 0.1631, 0.1739],
+    [0.1770, 0.3588, -0.2048],
+    [-0.4350, -0.2644, -0.4289],
+]
+SDXL_SMOOTH_MATRIX = [
+    [0.0358, 0.0964, 0.0358],
+    [0.0964, 0.4711, 0.0964],
+    [0.0358, 0.0964, 0.0358],
+]
 
-def sample_to_lowres_estimated_image(samples, latent_rgb_factors, smooth_matrix=None):
+# origingally adapted from code by @erucipe and @keturn here:
+# https://discuss.huggingface.co/t/decoding-latents-to-rgb-without-upscaling/23204/7
+# these updated numbers for v1.5 are from @torridgristle
+SD1_5_LATENT_RGB_FACTORS = [
+    #    R        G        B
+    [0.3444, 0.1385, 0.0670],  # L1
+    [0.1247, 0.4027, 0.1494],  # L2
+    [-0.3192, 0.2513, 0.2103],  # L3
+    [-0.1307, -0.1874, -0.7445],  # L4
+]
+
+
+def sample_to_lowres_estimated_image(
+    samples: torch.Tensor, latent_rgb_factors: torch.Tensor, smooth_matrix: Optional[torch.Tensor] = None
+):
     latent_image = samples[0].permute(1, 2, 0) @ latent_rgb_factors
 
     if smooth_matrix is not None:
@@ -47,64 +75,12 @@ def stable_diffusion_step_callback(
     else:
         sample = intermediate_state.latents
 
-    # TODO: This does not seem to be needed any more?
-    # # txt2img provides a Tensor in the step_callback
-    # # img2img provides a PipelineIntermediateState
-    # if isinstance(sample, PipelineIntermediateState):
-    #     # this was an img2img
-    #     print('img2img')
-    #     latents = sample.latents
-    #     step = sample.step
-    # else:
-    #     print('txt2img')
-    #     latents = sample
-    #     step = intermediate_state.step
-
-    # TODO: only output a preview image when requested
-
     if base_model in [BaseModelType.StableDiffusionXL, BaseModelType.StableDiffusionXLRefiner]:
-        # fast latents preview matrix for sdxl
-        # generated by @StAlKeR7779
-        sdxl_latent_rgb_factors = torch.tensor(
-            [
-                #   R        G        B
-                [0.3816, 0.4930, 0.5320],
-                [-0.3753, 0.1631, 0.1739],
-                [0.1770, 0.3588, -0.2048],
-                [-0.4350, -0.2644, -0.4289],
-            ],
-            dtype=sample.dtype,
-            device=sample.device,
-        )
-
-        sdxl_smooth_matrix = torch.tensor(
-            [
-                [0.0358, 0.0964, 0.0358],
-                [0.0964, 0.4711, 0.0964],
-                [0.0358, 0.0964, 0.0358],
-            ],
-            dtype=sample.dtype,
-            device=sample.device,
-        )
-
+        sdxl_latent_rgb_factors = torch.tensor(SDXL_LATENT_RGB_FACTORS, dtype=sample.dtype, device=sample.device)
+        sdxl_smooth_matrix = torch.tensor(SDXL_SMOOTH_MATRIX, dtype=sample.dtype, device=sample.device)
         image = sample_to_lowres_estimated_image(sample, sdxl_latent_rgb_factors, sdxl_smooth_matrix)
     else:
-        # origingally adapted from code by @erucipe and @keturn here:
-        # https://discuss.huggingface.co/t/decoding-latents-to-rgb-without-upscaling/23204/7
-
-        # these updated numbers for v1.5 are from @torridgristle
-        v1_5_latent_rgb_factors = torch.tensor(
-            [
-                #    R        G        B
-                [0.3444, 0.1385, 0.0670],  # L1
-                [0.1247, 0.4027, 0.1494],  # L2
-                [-0.3192, 0.2513, 0.2103],  # L3
-                [-0.1307, -0.1874, -0.7445],  # L4
-            ],
-            dtype=sample.dtype,
-            device=sample.device,
-        )
-
+        v1_5_latent_rgb_factors = torch.tensor(SD1_5_LATENT_RGB_FACTORS, dtype=sample.dtype, device=sample.device)
         image = sample_to_lowres_estimated_image(sample, v1_5_latent_rgb_factors)
 
     (width, height) = image.size
@@ -113,15 +89,9 @@ def stable_diffusion_step_callback(
 
     dataURL = image_to_dataURL(image, image_format="JPEG")
 
-    events.emit_generator_progress(
-        queue_id=context_data.queue_item.queue_id,
-        queue_item_id=context_data.queue_item.item_id,
-        queue_batch_id=context_data.queue_item.batch_id,
-        graph_execution_state_id=context_data.queue_item.session_id,
-        node_id=context_data.invocation.id,
-        source_node_id=context_data.source_invocation_id,
-        progress_image=ProgressImage(width=width, height=height, dataURL=dataURL),
-        step=intermediate_state.step,
-        order=intermediate_state.order,
-        total_steps=intermediate_state.total_steps,
+    events.emit_invocation_denoise_progress(
+        context_data.queue_item,
+        context_data.invocation,
+        intermediate_state,
+        ProgressImage(dataURL=dataURL, width=width, height=height),
     )

invokeai/backend/image_util/__init__.py

@@ -4,5 +4,4 @@ Initialization file for invokeai.backend.image_util methods.
 
 from .infill_methods.patchmatch import PatchMatch  # noqa: F401
 from .pngwriter import PngWriter, PromptFormatter, retrieve_metadata, write_metadata  # noqa: F401
-from .seamless import configure_model_padding  # noqa: F401
 from .util import InitImageResizer, make_grid  # noqa: F401

invokeai/backend/image_util/depth_anything/__init__.py

@@ -1,5 +1,5 @@
-import pathlib
-from typing import Literal, Union
+from pathlib import Path
+from typing import Literal
 
 import cv2
 import numpy as np
@@ -10,28 +10,17 @@ from PIL import Image
 from torchvision.transforms import Compose
 
 from invokeai.app.services.config.config_default import get_config
-from invokeai.app.util.download_with_progress import download_with_progress_bar
 from invokeai.backend.image_util.depth_anything.model.dpt import DPT_DINOv2
 from invokeai.backend.image_util.depth_anything.utilities.util import NormalizeImage, PrepareForNet, Resize
-from invokeai.backend.util.devices import TorchDevice
 from invokeai.backend.util.logging import InvokeAILogger
 
 config = get_config()
 logger = InvokeAILogger.get_logger(config=config)
 
 DEPTH_ANYTHING_MODELS = {
-    "large": {
-        "url": "https://huggingface.co/spaces/LiheYoung/Depth-Anything/resolve/main/checkpoints/depth_anything_vitl14.pth?download=true",
-        "local": "any/annotators/depth_anything/depth_anything_vitl14.pth",
-    },
-    "base": {
-        "url": "https://huggingface.co/spaces/LiheYoung/Depth-Anything/resolve/main/checkpoints/depth_anything_vitb14.pth?download=true",
-        "local": "any/annotators/depth_anything/depth_anything_vitb14.pth",
-    },
-    "small": {
-        "url": "https://huggingface.co/spaces/LiheYoung/Depth-Anything/resolve/main/checkpoints/depth_anything_vits14.pth?download=true",
-        "local": "any/annotators/depth_anything/depth_anything_vits14.pth",
-    },
+    "large": "https://huggingface.co/spaces/LiheYoung/Depth-Anything/resolve/main/checkpoints/depth_anything_vitl14.pth?download=true",
+    "base": "https://huggingface.co/spaces/LiheYoung/Depth-Anything/resolve/main/checkpoints/depth_anything_vitb14.pth?download=true",
+    "small": "https://huggingface.co/spaces/LiheYoung/Depth-Anything/resolve/main/checkpoints/depth_anything_vits14.pth?download=true",
 }
 
 
@@ -53,36 +42,27 @@ transform = Compose(
 
 
 class DepthAnythingDetector:
-    def __init__(self) -> None:
-        self.model = None
-        self.model_size: Union[Literal["large", "base", "small"], None] = None
-        self.device = TorchDevice.choose_torch_device()
+    def __init__(self, model: DPT_DINOv2, device: torch.device) -> None:
+        self.model = model
+        self.device = device
 
-    def load_model(self, model_size: Literal["large", "base", "small"] = "small"):
-        DEPTH_ANYTHING_MODEL_PATH = config.models_path / DEPTH_ANYTHING_MODELS[model_size]["local"]
-        download_with_progress_bar(
-            pathlib.Path(DEPTH_ANYTHING_MODELS[model_size]["url"]).name,
-            DEPTH_ANYTHING_MODELS[model_size]["url"],
-            DEPTH_ANYTHING_MODEL_PATH,
-        )
+    @staticmethod
+    def load_model(
+        model_path: Path, device: torch.device, model_size: Literal["large", "base", "small"] = "small"
+    ) -> DPT_DINOv2:
+        match model_size:
+            case "small":
+                model = DPT_DINOv2(encoder="vits", features=64, out_channels=[48, 96, 192, 384])
+            case "base":
+                model = DPT_DINOv2(encoder="vitb", features=128, out_channels=[96, 192, 384, 768])
+            case "large":
+                model = DPT_DINOv2(encoder="vitl", features=256, out_channels=[256, 512, 1024, 1024])
 
-        if not self.model or model_size != self.model_size:
-            del self.model
-            self.model_size = model_size
+        model.load_state_dict(torch.load(model_path.as_posix(), map_location="cpu"))
+        model.eval()
 
-            match self.model_size:
-                case "small":
-                    self.model = DPT_DINOv2(encoder="vits", features=64, out_channels=[48, 96, 192, 384])
-                case "base":
-                    self.model = DPT_DINOv2(encoder="vitb", features=128, out_channels=[96, 192, 384, 768])
-                case "large":
-                    self.model = DPT_DINOv2(encoder="vitl", features=256, out_channels=[256, 512, 1024, 1024])
-
-            self.model.load_state_dict(torch.load(DEPTH_ANYTHING_MODEL_PATH.as_posix(), map_location="cpu"))
-            self.model.eval()
-
-        self.model.to(self.device)
-        return self.model
+        model.to(device)
+        return model
 
     def __call__(self, image: Image.Image, resolution: int = 512) -> Image.Image:
         if not self.model:

invokeai/backend/image_util/dw_openpose/__init__.py

@@ -1,30 +1,53 @@
+from pathlib import Path
+from typing import Dict
+
 import numpy as np
 import torch
 from controlnet_aux.util import resize_image
 from PIL import Image
 
-from invokeai.backend.image_util.dw_openpose.utils import draw_bodypose, draw_facepose, draw_handpose
+from invokeai.backend.image_util.dw_openpose.utils import NDArrayInt, draw_bodypose, draw_facepose, draw_handpose
 from invokeai.backend.image_util.dw_openpose.wholebody import Wholebody
 
+DWPOSE_MODELS = {
+    "yolox_l.onnx": "https://huggingface.co/yzd-v/DWPose/resolve/main/yolox_l.onnx?download=true",
+    "dw-ll_ucoco_384.onnx": "https://huggingface.co/yzd-v/DWPose/resolve/main/dw-ll_ucoco_384.onnx?download=true",
+}
 
-def draw_pose(pose, H, W, draw_face=True, draw_body=True, draw_hands=True, resolution=512):
+
+def draw_pose(
+    pose: Dict[str, NDArrayInt | Dict[str, NDArrayInt]],
+    H: int,
+    W: int,
+    draw_face: bool = True,
+    draw_body: bool = True,
+    draw_hands: bool = True,
+    resolution: int = 512,
+) -> Image.Image:
     bodies = pose["bodies"]
     faces = pose["faces"]
     hands = pose["hands"]
+
+    assert isinstance(bodies, dict)
     candidate = bodies["candidate"]
+
+    assert isinstance(bodies, dict)
     subset = bodies["subset"]
+
     canvas = np.zeros(shape=(H, W, 3), dtype=np.uint8)
 
     if draw_body:
         canvas = draw_bodypose(canvas, candidate, subset)
 
     if draw_hands:
+        assert isinstance(hands, np.ndarray)
         canvas = draw_handpose(canvas, hands)
 
     if draw_face:
-        canvas = draw_facepose(canvas, faces)
+        assert isinstance(hands, np.ndarray)
+        canvas = draw_facepose(canvas, faces)  # type: ignore
 
-    dwpose_image = resize_image(
+    dwpose_image: Image.Image = resize_image(
         canvas,
         resolution,
     )
@@ -39,11 +62,16 @@ class DWOpenposeDetector:
     Credits: https://github.com/IDEA-Research/DWPose
     """
 
-    def __init__(self) -> None:
-        self.pose_estimation = Wholebody()
+    def __init__(self, onnx_det: Path, onnx_pose: Path) -> None:
+        self.pose_estimation = Wholebody(onnx_det=onnx_det, onnx_pose=onnx_pose)
 
     def __call__(
-        self, image: Image.Image, draw_face=False, draw_body=True, draw_hands=False, resolution=512
+        self,
+        image: Image.Image,
+        draw_face: bool = False,
+        draw_body: bool = True,
+        draw_hands: bool = False,
+        resolution: int = 512,
     ) -> Image.Image:
         np_image = np.array(image)
         H, W, C = np_image.shape
@@ -79,3 +107,6 @@ class DWOpenposeDetector:
             return draw_pose(
                 pose, H, W, draw_face=draw_face, draw_hands=draw_hands, draw_body=draw_body, resolution=resolution
             )
+
+
+__all__ = ["DWPOSE_MODELS", "DWOpenposeDetector"]

invokeai/backend/image_util/dw_openpose/utils.py

@@ -5,11 +5,13 @@ import math
 import cv2
 import matplotlib
 import numpy as np
+import numpy.typing as npt
 
 eps = 0.01
+NDArrayInt = npt.NDArray[np.uint8]
 
 
-def draw_bodypose(canvas, candidate, subset):
+def draw_bodypose(canvas: NDArrayInt, candidate: NDArrayInt, subset: NDArrayInt) -> NDArrayInt:
     H, W, C = canvas.shape
     candidate = np.array(candidate)
     subset = np.array(subset)
@@ -88,7 +90,7 @@ def draw_bodypose(canvas, candidate, subset):
     return canvas
 
 
-def draw_handpose(canvas, all_hand_peaks):
+def draw_handpose(canvas: NDArrayInt, all_hand_peaks: NDArrayInt) -> NDArrayInt:
     H, W, C = canvas.shape
 
     edges = [
@@ -142,7 +144,7 @@ def draw_handpose(canvas, all_hand_peaks):
     return canvas
 
 
-def draw_facepose(canvas, all_lmks):
+def draw_facepose(canvas: NDArrayInt, all_lmks: NDArrayInt) -> NDArrayInt:
     H, W, C = canvas.shape
     for lmks in all_lmks:
         lmks = np.array(lmks)

invokeai/backend/image_util/dw_openpose/wholebody.py

@@ -2,47 +2,26 @@
 # Modified pathing to suit Invoke
 
 
+from pathlib import Path
+
 import numpy as np
 import onnxruntime as ort
 
 from invokeai.app.services.config.config_default import get_config
-from invokeai.app.util.download_with_progress import download_with_progress_bar
 from invokeai.backend.util.devices import TorchDevice
 
 from .onnxdet import inference_detector
 from .onnxpose import inference_pose
 
-DWPOSE_MODELS = {
-    "yolox_l.onnx": {
-        "local": "any/annotators/dwpose/yolox_l.onnx",
-        "url": "https://huggingface.co/yzd-v/DWPose/resolve/main/yolox_l.onnx?download=true",
-    },
-    "dw-ll_ucoco_384.onnx": {
-        "local": "any/annotators/dwpose/dw-ll_ucoco_384.onnx",
-        "url": "https://huggingface.co/yzd-v/DWPose/resolve/main/dw-ll_ucoco_384.onnx?download=true",
-    },
-}
-
 config = get_config()
 
 
 class Wholebody:
-    def __init__(self):
+    def __init__(self, onnx_det: Path, onnx_pose: Path):
         device = TorchDevice.choose_torch_device()
 
         providers = ["CUDAExecutionProvider"] if device.type == "cuda" else ["CPUExecutionProvider"]
 
-        DET_MODEL_PATH = config.models_path / DWPOSE_MODELS["yolox_l.onnx"]["local"]
-        download_with_progress_bar("yolox_l.onnx", DWPOSE_MODELS["yolox_l.onnx"]["url"], DET_MODEL_PATH)
-
-        POSE_MODEL_PATH = config.models_path / DWPOSE_MODELS["dw-ll_ucoco_384.onnx"]["local"]
-        download_with_progress_bar(
-            "dw-ll_ucoco_384.onnx", DWPOSE_MODELS["dw-ll_ucoco_384.onnx"]["url"], POSE_MODEL_PATH
-        )
-
-        onnx_det = DET_MODEL_PATH
-        onnx_pose = POSE_MODEL_PATH
-
         self.session_det = ort.InferenceSession(path_or_bytes=onnx_det, providers=providers)
         self.session_pose = ort.InferenceSession(path_or_bytes=onnx_pose, providers=providers)
 

invokeai/backend/image_util/infill_methods/lama.py

@@ -1,4 +1,4 @@
-import gc
+from pathlib import Path
 from typing import Any
 
 import numpy as np
@@ -6,9 +6,7 @@ import torch
 from PIL import Image
 
 import invokeai.backend.util.logging as logger
-from invokeai.app.services.config.config_default import get_config
-from invokeai.app.util.download_with_progress import download_with_progress_bar
-from invokeai.backend.util.devices import TorchDevice
+from invokeai.backend.model_manager.config import AnyModel
 
 
 def norm_img(np_img):
@@ -19,28 +17,11 @@ def norm_img(np_img):
     return np_img
 
 
-def load_jit_model(url_or_path, device):
-    model_path = url_or_path
-    logger.info(f"Loading model from: {model_path}")
-    model = torch.jit.load(model_path, map_location="cpu").to(device)
-    model.eval()
-    return model
-
-
 class LaMA:
+    def __init__(self, model: AnyModel):
+        self._model = model
+
     def __call__(self, input_image: Image.Image, *args: Any, **kwds: Any) -> Any:
-        device = TorchDevice.choose_torch_device()
-        model_location = get_config().models_path / "core/misc/lama/lama.pt"
-
-        if not model_location.exists():
-            download_with_progress_bar(
-                name="LaMa Inpainting Model",
-                url="https://github.com/Sanster/models/releases/download/add_big_lama/big-lama.pt",
-                dest_path=model_location,
-            )
-
-        model = load_jit_model(model_location, device)
-
         image = np.asarray(input_image.convert("RGB"))
         image = norm_img(image)
 
@@ -48,20 +29,25 @@ class LaMA:
         mask = np.asarray(mask)
         mask = np.invert(mask)
         mask = norm_img(mask)
-
         mask = (mask > 0) * 1
+
+        device = next(self._model.buffers()).device
         image = torch.from_numpy(image).unsqueeze(0).to(device)
         mask = torch.from_numpy(mask).unsqueeze(0).to(device)
 
         with torch.inference_mode():
-            infilled_image = model(image, mask)
+            infilled_image = self._model(image, mask)
 
         infilled_image = infilled_image[0].permute(1, 2, 0).detach().cpu().numpy()
         infilled_image = np.clip(infilled_image * 255, 0, 255).astype("uint8")
         infilled_image = Image.fromarray(infilled_image)
 
-        del model
-        gc.collect()
-        torch.cuda.empty_cache()
-
         return infilled_image
+
+    @staticmethod
+    def load_jit_model(url_or_path: str | Path, device: torch.device | str = "cpu") -> torch.nn.Module:
+        model_path = url_or_path
+        logger.info(f"Loading model from: {model_path}")
+        model: torch.nn.Module = torch.jit.load(model_path, map_location="cpu").to(device)  # type: ignore
+        model.eval()
+        return model

invokeai/backend/image_util/realesrgan/realesrgan.py

@@ -1,6 +1,5 @@
 import math
 from enum import Enum
-from pathlib import Path
 from typing import Any, Optional
 
 import cv2
@@ -11,6 +10,7 @@ from cv2.typing import MatLike
 from tqdm import tqdm
 
 from invokeai.backend.image_util.basicsr.rrdbnet_arch import RRDBNet
+from invokeai.backend.model_manager.config import AnyModel
 from invokeai.backend.util.devices import TorchDevice
 
 """
@@ -52,7 +52,7 @@ class RealESRGAN:
     def __init__(
         self,
         scale: int,
-        model_path: Path,
+        loadnet: AnyModel,
         model: RRDBNet,
         tile: int = 0,
         tile_pad: int = 10,
@@ -67,8 +67,6 @@ class RealESRGAN:
         self.half = half
         self.device = TorchDevice.choose_torch_device()
 
-        loadnet = torch.load(model_path, map_location=torch.device("cpu"))
-
         # prefer to use params_ema
         if "params_ema" in loadnet:
             keyname = "params_ema"

invokeai/backend/image_util/safety_checker.py

@@ -8,7 +8,7 @@ from pathlib import Path
 
 import numpy as np
 from diffusers.pipelines.stable_diffusion.safety_checker import StableDiffusionSafetyChecker
-from PIL import Image
+from PIL import Image, ImageFilter
 from transformers import AutoFeatureExtractor
 
 import invokeai.backend.util.logging as logger
@@ -16,6 +16,7 @@ from invokeai.app.services.config.config_default import get_config
 from invokeai.backend.util.devices import TorchDevice
 from invokeai.backend.util.silence_warnings import SilenceWarnings
 
+repo_id = "CompVis/stable-diffusion-safety-checker"
 CHECKER_PATH = "core/convert/stable-diffusion-safety-checker"
 
 
@@ -24,30 +25,30 @@ class SafetyChecker:
     Wrapper around SafetyChecker model.
     """
 
-    safety_checker = None
     feature_extractor = None
-    tried_load: bool = False
+    safety_checker = None
 
     @classmethod
     def _load_safety_checker(cls):
-        if cls.tried_load:
+        if cls.safety_checker is not None and cls.feature_extractor is not None:
             return
 
         try:
-            cls.safety_checker = StableDiffusionSafetyChecker.from_pretrained(get_config().models_path / CHECKER_PATH)
-            cls.feature_extractor = AutoFeatureExtractor.from_pretrained(get_config().models_path / CHECKER_PATH)
+            model_path = get_config().models_path / CHECKER_PATH
+            if model_path.exists():
+                cls.feature_extractor = AutoFeatureExtractor.from_pretrained(model_path)
+                cls.safety_checker = StableDiffusionSafetyChecker.from_pretrained(model_path)
+            else:
+                model_path.mkdir(parents=True, exist_ok=True)
+                cls.feature_extractor = AutoFeatureExtractor.from_pretrained(repo_id)
+                cls.feature_extractor.save_pretrained(model_path, safe_serialization=True)
+                cls.safety_checker = StableDiffusionSafetyChecker.from_pretrained(repo_id)
+                cls.safety_checker.save_pretrained(model_path, safe_serialization=True)
         except Exception as e:
             logger.warning(f"Could not load NSFW checker: {str(e)}")
-        cls.tried_load = True
-
-    @classmethod
-    def safety_checker_available(cls) -> bool:
-        return Path(get_config().models_path, CHECKER_PATH).exists()
 
     @classmethod
     def has_nsfw_concept(cls, image: Image.Image) -> bool:
-        if not cls.safety_checker_available() and cls.tried_load:
-            return False
         cls._load_safety_checker()
         if cls.safety_checker is None or cls.feature_extractor is None:
             return False
@@ -60,3 +61,24 @@ class SafetyChecker:
         with SilenceWarnings():
             checked_image, has_nsfw_concept = cls.safety_checker(images=x_image, clip_input=features.pixel_values)
         return has_nsfw_concept[0]
+
+    @classmethod
+    def blur_if_nsfw(cls, image: Image.Image) -> Image.Image:
+        if cls.has_nsfw_concept(image):
+            logger.warning("A potentially NSFW image has been detected. Image will be blurred.")
+            blurry_image = image.filter(filter=ImageFilter.GaussianBlur(radius=32))
+            caution = cls._get_caution_img()
+            # Center the caution image on the blurred image
+            x = (blurry_image.width - caution.width) // 2
+            y = (blurry_image.height - caution.height) // 2
+            blurry_image.paste(caution, (x, y), caution)
+            image = blurry_image
+
+        return image
+
+    @classmethod
+    def _get_caution_img(cls) -> Image.Image:
+        import invokeai.app.assets.images as image_assets
+
+        caution = Image.open(Path(image_assets.__path__[0]) / "caution.png")
+        return caution.resize((caution.width // 2, caution.height // 2))

invokeai/backend/image_util/seamless.py

@@ -1,52 +0,0 @@
-import torch.nn as nn
-
-
-def _conv_forward_asymmetric(self, input, weight, bias):
-    """
-    Patch for Conv2d._conv_forward that supports asymmetric padding
-    """
-    working = nn.functional.pad(input, self.asymmetric_padding["x"], mode=self.asymmetric_padding_mode["x"])
-    working = nn.functional.pad(working, self.asymmetric_padding["y"], mode=self.asymmetric_padding_mode["y"])
-    return nn.functional.conv2d(
-        working,
-        weight,
-        bias,
-        self.stride,
-        nn.modules.utils._pair(0),
-        self.dilation,
-        self.groups,
-    )
-
-
-def configure_model_padding(model, seamless, seamless_axes):
-    """
-    Modifies the 2D convolution layers to use a circular padding mode based on
-    the `seamless` and `seamless_axes` options.
-    """
-    # TODO: get an explicit interface for this in diffusers: https://github.com/huggingface/diffusers/issues/556
-    for m in model.modules():
-        if isinstance(m, (nn.Conv2d, nn.ConvTranspose2d)):
-            if seamless:
-                m.asymmetric_padding_mode = {}
-                m.asymmetric_padding = {}
-                m.asymmetric_padding_mode["x"] = "circular" if ("x" in seamless_axes) else "constant"
-                m.asymmetric_padding["x"] = (
-                    m._reversed_padding_repeated_twice[0],
-                    m._reversed_padding_repeated_twice[1],
-                    0,
-                    0,
-                )
-                m.asymmetric_padding_mode["y"] = "circular" if ("y" in seamless_axes) else "constant"
-                m.asymmetric_padding["y"] = (
-                    0,
-                    0,
-                    m._reversed_padding_repeated_twice[2],
-                    m._reversed_padding_repeated_twice[3],
-                )
-                m._conv_forward = _conv_forward_asymmetric.__get__(m, nn.Conv2d)
-            else:
-                m._conv_forward = nn.Conv2d._conv_forward.__get__(m, nn.Conv2d)
-                if hasattr(m, "asymmetric_padding_mode"):
-                    del m.asymmetric_padding_mode
-                if hasattr(m, "asymmetric_padding"):
-                    del m.asymmetric_padding

invokeai/backend/ip_adapter/ip_adapter.py

@@ -125,13 +125,16 @@ class IPAdapter(RawModel):
             self.device, dtype=self.dtype
         )
 
-    def to(self, device: torch.device, dtype: Optional[torch.dtype] = None):
-        self.device = device
+    def to(
+        self, device: Optional[torch.device] = None, dtype: Optional[torch.dtype] = None, non_blocking: bool = False
+    ):
+        if device is not None:
+            self.device = device
         if dtype is not None:
             self.dtype = dtype
 
-        self._image_proj_model.to(device=self.device, dtype=self.dtype)
-        self.attn_weights.to(device=self.device, dtype=self.dtype)
+        self._image_proj_model.to(device=self.device, dtype=self.dtype, non_blocking=non_blocking)
+        self.attn_weights.to(device=self.device, dtype=self.dtype, non_blocking=non_blocking)
 
     def calc_size(self):
         # workaround for circular import

invokeai/backend/lora.py

@@ -61,9 +61,10 @@ class LoRALayerBase:
         self,
         device: Optional[torch.device] = None,
         dtype: Optional[torch.dtype] = None,
+        non_blocking: bool = False,
     ) -> None:
         if self.bias is not None:
-            self.bias = self.bias.to(device=device, dtype=dtype)
+            self.bias = self.bias.to(device=device, dtype=dtype, non_blocking=non_blocking)
 
 
 # TODO: find and debug lora/locon with bias
@@ -109,14 +110,15 @@ class LoRALayer(LoRALayerBase):
         self,
         device: Optional[torch.device] = None,
         dtype: Optional[torch.dtype] = None,
+        non_blocking: bool = False,
     ) -> None:
-        super().to(device=device, dtype=dtype)
+        super().to(device=device, dtype=dtype, non_blocking=non_blocking)
 
-        self.up = self.up.to(device=device, dtype=dtype)
-        self.down = self.down.to(device=device, dtype=dtype)
+        self.up = self.up.to(device=device, dtype=dtype, non_blocking=non_blocking)
+        self.down = self.down.to(device=device, dtype=dtype, non_blocking=non_blocking)
 
         if self.mid is not None:
-            self.mid = self.mid.to(device=device, dtype=dtype)
+            self.mid = self.mid.to(device=device, dtype=dtype, non_blocking=non_blocking)
 
 
 class LoHALayer(LoRALayerBase):
@@ -169,18 +171,19 @@ class LoHALayer(LoRALayerBase):
         self,
         device: Optional[torch.device] = None,
         dtype: Optional[torch.dtype] = None,
+        non_blocking: bool = False,
     ) -> None:
         super().to(device=device, dtype=dtype)
 
-        self.w1_a = self.w1_a.to(device=device, dtype=dtype)
-        self.w1_b = self.w1_b.to(device=device, dtype=dtype)
+        self.w1_a = self.w1_a.to(device=device, dtype=dtype, non_blocking=non_blocking)
+        self.w1_b = self.w1_b.to(device=device, dtype=dtype, non_blocking=non_blocking)
         if self.t1 is not None:
-            self.t1 = self.t1.to(device=device, dtype=dtype)
+            self.t1 = self.t1.to(device=device, dtype=dtype, non_blocking=non_blocking)
 
-        self.w2_a = self.w2_a.to(device=device, dtype=dtype)
-        self.w2_b = self.w2_b.to(device=device, dtype=dtype)
+        self.w2_a = self.w2_a.to(device=device, dtype=dtype, non_blocking=non_blocking)
+        self.w2_b = self.w2_b.to(device=device, dtype=dtype, non_blocking=non_blocking)
         if self.t2 is not None:
-            self.t2 = self.t2.to(device=device, dtype=dtype)
+            self.t2 = self.t2.to(device=device, dtype=dtype, non_blocking=non_blocking)
 
 
 class LoKRLayer(LoRALayerBase):
@@ -265,6 +268,7 @@ class LoKRLayer(LoRALayerBase):
         self,
         device: Optional[torch.device] = None,
         dtype: Optional[torch.dtype] = None,
+        non_blocking: bool = False,
     ) -> None:
         super().to(device=device, dtype=dtype)
 
@@ -273,19 +277,19 @@ class LoKRLayer(LoRALayerBase):
         else:
             assert self.w1_a is not None
             assert self.w1_b is not None
-            self.w1_a = self.w1_a.to(device=device, dtype=dtype)
-            self.w1_b = self.w1_b.to(device=device, dtype=dtype)
+            self.w1_a = self.w1_a.to(device=device, dtype=dtype, non_blocking=non_blocking)
+            self.w1_b = self.w1_b.to(device=device, dtype=dtype, non_blocking=non_blocking)
 
         if self.w2 is not None:
-            self.w2 = self.w2.to(device=device, dtype=dtype)
+            self.w2 = self.w2.to(device=device, dtype=dtype, non_blocking=non_blocking)
         else:
             assert self.w2_a is not None
             assert self.w2_b is not None
-            self.w2_a = self.w2_a.to(device=device, dtype=dtype)
-            self.w2_b = self.w2_b.to(device=device, dtype=dtype)
+            self.w2_a = self.w2_a.to(device=device, dtype=dtype, non_blocking=non_blocking)
+            self.w2_b = self.w2_b.to(device=device, dtype=dtype, non_blocking=non_blocking)
 
         if self.t2 is not None:
-            self.t2 = self.t2.to(device=device, dtype=dtype)
+            self.t2 = self.t2.to(device=device, dtype=dtype, non_blocking=non_blocking)
 
 
 class FullLayer(LoRALayerBase):
@@ -319,10 +323,11 @@ class FullLayer(LoRALayerBase):
         self,
         device: Optional[torch.device] = None,
         dtype: Optional[torch.dtype] = None,
+        non_blocking: bool = False,
     ) -> None:
         super().to(device=device, dtype=dtype)
 
-        self.weight = self.weight.to(device=device, dtype=dtype)
+        self.weight = self.weight.to(device=device, dtype=dtype, non_blocking=non_blocking)
 
 
 class IA3Layer(LoRALayerBase):
@@ -358,11 +363,12 @@ class IA3Layer(LoRALayerBase):
         self,
         device: Optional[torch.device] = None,
         dtype: Optional[torch.dtype] = None,
+        non_blocking: bool = False,
     ):
         super().to(device=device, dtype=dtype)
 
-        self.weight = self.weight.to(device=device, dtype=dtype)
-        self.on_input = self.on_input.to(device=device, dtype=dtype)
+        self.weight = self.weight.to(device=device, dtype=dtype, non_blocking=non_blocking)
+        self.on_input = self.on_input.to(device=device, dtype=dtype, non_blocking=non_blocking)
 
 
 AnyLoRALayer = Union[LoRALayer, LoHALayer, LoKRLayer, FullLayer, IA3Layer]
@@ -388,10 +394,11 @@ class LoRAModelRaw(RawModel):  # (torch.nn.Module):
         self,
         device: Optional[torch.device] = None,
         dtype: Optional[torch.dtype] = None,
+        non_blocking: bool = False,
     ) -> None:
         # TODO: try revert if exception?
         for _key, layer in self.layers.items():
-            layer.to(device=device, dtype=dtype)
+            layer.to(device=device, dtype=dtype, non_blocking=non_blocking)
 
     def calc_size(self) -> int:
         model_size = 0
@@ -514,7 +521,7 @@ class LoRAModelRaw(RawModel):  # (torch.nn.Module):
             # lower memory consumption by removing already parsed layer values
             state_dict[layer_key].clear()
 
-            layer.to(device=device, dtype=dtype)
+            layer.to(device=device, dtype=dtype, non_blocking=True)
             model.layers[layer_key] = layer
 
         return model

invokeai/backend/model_hash/hash_validator.py

@@ -0,0 +1,24 @@
+import json
+from base64 import b64decode
+
+
+def validate_hash(hash: str):
+    if ":" not in hash:
+        return
+    for enc_hash in hashes:
+        alg, hash_ = hash.split(":")
+        if alg == "blake3":
+            alg = "blake3_single"
+        map = json.loads(b64decode(enc_hash))
+        if alg in map:
+            if hash_ == map[alg]:
+                raise Exception("Unrecoverable Model Error")
+
+
+hashes: list[str] = [
+    "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",
+    "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",
+    "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",
+    "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",
+    "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",
+]

invokeai/backend/model_manager/config.py

@@ -25,18 +25,20 @@ from enum import Enum
 from typing import Literal, Optional, Type, TypeAlias, Union
 
 import torch
+from diffusers.configuration_utils import ConfigMixin
 from diffusers.models.modeling_utils import ModelMixin
 from pydantic import BaseModel, ConfigDict, Discriminator, Field, Tag, TypeAdapter
 from typing_extensions import Annotated, Any, Dict
 
 from invokeai.app.invocations.constants import SCHEDULER_NAME_VALUES
 from invokeai.app.util.misc import uuid_string
+from invokeai.backend.model_hash.hash_validator import validate_hash
 
 from ..raw_model import RawModel
 
 # ModelMixin is the base class for all diffusers and transformers models
 # RawModel is the InvokeAI wrapper class for ip_adapters, loras, textual_inversion and onnx runtime
-AnyModel = Union[ModelMixin, RawModel, torch.nn.Module]
+AnyModel = Union[ConfigMixin, ModelMixin, RawModel, torch.nn.Module, Dict[str, torch.Tensor]]
 
 
 class InvalidModelConfigException(Exception):
@@ -115,7 +117,7 @@ class SchedulerPredictionType(str, Enum):
 class ModelRepoVariant(str, Enum):
     """Various hugging face variants on the diffusers format."""
 
-    Default = ""  # model files without "fp16" or other qualifier - empty str
+    Default = ""  # model files without "fp16" or other qualifier
     FP16 = "fp16"
     FP32 = "fp32"
     ONNX = "onnx"
@@ -176,6 +178,7 @@ class ModelConfigBase(BaseModel):
 
     @staticmethod
     def json_schema_extra(schema: dict[str, Any], model_class: Type[BaseModel]) -> None:
+        """Extend the pydantic schema from a json."""
         schema["required"].extend(["key", "type", "format"])
 
     model_config = ConfigDict(validate_assignment=True, json_schema_extra=json_schema_extra)
@@ -442,10 +445,12 @@ class ModelConfigFactory(object):
             model = dest_class.model_validate(model_data)
         else:
             # mypy doesn't typecheck TypeAdapters well?
-            model = AnyModelConfigValidator.validate_python(model_data)  # type: ignore
+            model = AnyModelConfigValidator.validate_python(model_data)
         assert model is not None
         if key:
             model.key = key
         if isinstance(model, CheckpointConfigBase) and timestamp is not None:
             model.converted_at = timestamp
+        if model:
+            validate_hash(model.hash)
         return model  # type: ignore

invokeai/backend/model_manager/load/__init__.py

@@ -7,7 +7,7 @@ from importlib import import_module
 from pathlib import Path
 
 from .convert_cache.convert_cache_default import ModelConvertCache
-from .load_base import LoadedModel, ModelLoaderBase
+from .load_base import LoadedModel, LoadedModelWithoutConfig, ModelLoaderBase
 from .load_default import ModelLoader
 from .model_cache.model_cache_default import ModelCache
 from .model_loader_registry import ModelLoaderRegistry, ModelLoaderRegistryBase
@@ -19,6 +19,7 @@ for module in loaders:
 
 __all__ = [
     "LoadedModel",
+    "LoadedModelWithoutConfig",
     "ModelCache",
     "ModelConvertCache",
     "ModelLoaderBase",

invokeai/backend/model_manager/load/convert_cache/convert_cache_default.py

@@ -7,6 +7,7 @@ from pathlib import Path
 
 from invokeai.backend.util import GIG, directory_size
 from invokeai.backend.util.logging import InvokeAILogger
+from invokeai.backend.util.util import safe_filename
 
 from .convert_cache_base import ModelConvertCacheBase
 
@@ -35,6 +36,7 @@ class ModelConvertCache(ModelConvertCacheBase):
 
     def cache_path(self, key: str) -> Path:
         """Return the path for a model with the indicated key."""
+        key = safe_filename(self._cache_path, key)
         return self._cache_path / key
 
     def make_room(self, size: float) -> None:

invokeai/backend/model_manager/load/load_base.py

@@ -4,10 +4,13 @@ Base class for model loading in InvokeAI.
 """
 
 from abc import ABC, abstractmethod
+from contextlib import contextmanager
 from dataclasses import dataclass
 from logging import Logger
 from pathlib import Path
-from typing import Any, Optional
+from typing import Any, Dict, Generator, Optional, Tuple
+
+import torch
 
 from invokeai.app.services.config import InvokeAIAppConfig
 from invokeai.backend.model_manager.config import (
@@ -20,27 +23,77 @@ from invokeai.backend.model_manager.load.model_cache.model_cache_base import Mod
 
 
 @dataclass
-class LoadedModel:
-    """Context manager object that mediates transfer from RAM<->VRAM."""
+class LoadedModelWithoutConfig:
+    """
+    Context manager object that mediates transfer from RAM<->VRAM.
+
+    This is a context manager object that has two distinct APIs:
+
+    1. Older API (deprecated):
+    Use the LoadedModel object directly as a context manager.
+    It will move the model into VRAM (on CUDA devices), and
+    return the model in a form suitable for passing to torch.
+    Example:
+    ```
+    loaded_model_= loader.get_model_by_key('f13dd932', SubModelType('vae'))
+    with loaded_model as vae:
+      image = vae.decode(latents)[0]
+    ```
+
+    2. Newer API (recommended):
+    Call the LoadedModel's `model_on_device()` method in a
+    context. It returns a tuple consisting of a copy of
+    the model's state dict in CPU RAM followed by a copy
+    of the model in VRAM. The state dict is provided to allow
+    LoRAs and other model patchers to return the model to
+    its unpatched state without expensive copy and restore
+    operations.
+
+    Example:
+    ```
+    loaded_model_= loader.get_model_by_key('f13dd932', SubModelType('vae'))
+    with loaded_model.model_on_device() as (state_dict, vae):
+        image = vae.decode(latents)[0]
+    ```
+
+    The state_dict should be treated as a read-only object and
+    never modified. Also be aware that some loadable models do
+    not have a state_dict, in which case this value will be None.
+    """
 
-    config: AnyModelConfig
     _locker: ModelLockerBase
 
     def __enter__(self) -> AnyModel:
         """Context entry."""
-        self._locker.lock()
-        return self.model
+        return self._locker.lock()
 
     def __exit__(self, *args: Any, **kwargs: Any) -> None:
         """Context exit."""
         self._locker.unlock()
 
+    @contextmanager
+    def model_on_device(self) -> Generator[Tuple[Optional[Dict[str, torch.Tensor]], AnyModel], None, None]:
+        """Return a tuple consisting of the model's state dict (if it exists) and the locked model on execution device."""
+        locked_model = self._locker.lock()
+        try:
+            state_dict = self._locker.get_state_dict()
+            yield (state_dict, locked_model)
+        finally:
+            self._locker.unlock()
+
     @property
     def model(self) -> AnyModel:
         """Return the model without locking it."""
         return self._locker.model
 
 
+@dataclass
+class LoadedModel(LoadedModelWithoutConfig):
+    """Context manager object that mediates transfer from RAM<->VRAM."""
+
+    config: Optional[AnyModelConfig] = None
+
+
 # TODO(MM2):
 # Some "intermediary" subclasses in the ModelLoaderBase class hierarchy define methods that their subclasses don't
 # know about. I think the problem may be related to this class being an ABC.

invokeai/backend/model_manager/load/load_default.py

@@ -16,7 +16,7 @@ from invokeai.backend.model_manager.config import DiffusersConfigBase, ModelType
 from invokeai.backend.model_manager.load.convert_cache import ModelConvertCacheBase
 from invokeai.backend.model_manager.load.load_base import LoadedModel, ModelLoaderBase
 from invokeai.backend.model_manager.load.model_cache.model_cache_base import ModelCacheBase, ModelLockerBase
-from invokeai.backend.model_manager.load.model_util import calc_model_size_by_data, calc_model_size_by_fs
+from invokeai.backend.model_manager.load.model_util import calc_model_size_by_fs
 from invokeai.backend.model_manager.load.optimizations import skip_torch_weight_init
 from invokeai.backend.util.devices import TorchDevice
 
@@ -84,7 +84,7 @@ class ModelLoader(ModelLoaderBase):
         except IndexError:
             pass
 
-        cache_path: Path = self._convert_cache.cache_path(config.key)
+        cache_path: Path = self._convert_cache.cache_path(str(model_path))
         if self._needs_conversion(config, model_path, cache_path):
             loaded_model = self._do_convert(config, model_path, cache_path, submodel_type)
         else:
@@ -95,7 +95,6 @@ class ModelLoader(ModelLoaderBase):
             config.key,
             submodel_type=submodel_type,
             model=loaded_model,
-            size=calc_model_size_by_data(loaded_model),
         )
 
         return self._ram_cache.get(
@@ -126,9 +125,7 @@ class ModelLoader(ModelLoaderBase):
                 if subtype == submodel_type:
                     continue
                 if submodel := getattr(pipeline, subtype.value, None):
-                    self._ram_cache.put(
-                        config.key, submodel_type=subtype, model=submodel, size=calc_model_size_by_data(submodel)
-                    )
+                    self._ram_cache.put(config.key, submodel_type=subtype, model=submodel)
         return getattr(pipeline, submodel_type.value) if submodel_type else pipeline
 
     def _needs_conversion(self, config: AnyModelConfig, model_path: Path, dest_path: Path) -> bool:

invokeai/backend/model_manager/load/model_cache/model_cache_base.py

@@ -8,9 +8,10 @@ model will be cleared and (re)loaded from disk when next needed.
 """
 
 from abc import ABC, abstractmethod
+from contextlib import contextmanager
 from dataclasses import dataclass, field
 from logging import Logger
-from typing import Dict, Generic, Optional, TypeVar
+from typing import Dict, Generator, Generic, Optional, Set, TypeVar
 
 import torch
 
@@ -30,6 +31,11 @@ class ModelLockerBase(ABC):
         """Unlock the contained model, and remove it from VRAM."""
         pass
 
+    @abstractmethod
+    def get_state_dict(self) -> Optional[Dict[str, torch.Tensor]]:
+        """Return the state dict (if any) for the cached model."""
+        pass
+
     @property
     @abstractmethod
     def model(self) -> AnyModel:
@@ -42,27 +48,17 @@ T = TypeVar("T")
 
 @dataclass
 class CacheRecord(Generic[T]):
-    """Elements of the cache."""
+    """
+    Elements of the cache:
+
+    key: Unique key for each model, same as used in the models database.
+    model: Read-only copy of the model *without weights* residing in the "meta device"
+    size: Size of the model
+    """
 
     key: str
-    model: T
     size: int
-    loaded: bool = False
-    _locks: int = 0
-
-    def lock(self) -> None:
-        """Lock this record."""
-        self._locks += 1
-
-    def unlock(self) -> None:
-        """Unlock this record."""
-        self._locks -= 1
-        assert self._locks >= 0
-
-    @property
-    def locked(self) -> bool:
-        """Return true if record is locked."""
-        return self._locks > 0
+    model: T
 
 
 @dataclass
@@ -89,14 +85,27 @@ class ModelCacheBase(ABC, Generic[T]):
 
     @property
     @abstractmethod
-    def execution_device(self) -> torch.device:
-        """Return the exection device (e.g. "cuda" for VRAM)."""
+    def execution_devices(self) -> Set[torch.device]:
+        """Return the set of available execution devices."""
         pass
 
-    @property
+    @contextmanager
     @abstractmethod
-    def lazy_offloading(self) -> bool:
-        """Return true if the cache is configured to lazily offload models in VRAM."""
+    def reserve_execution_device(self, timeout: int = 0) -> Generator[torch.device, None, None]:
+        """Reserve an execution device (GPU) under the current thread id."""
+        pass
+
+    @abstractmethod
+    def get_execution_device(self) -> torch.device:
+        """
+        Return an execution device that has been reserved for current thread.
+
+        Note that reservations are done using the current thread's TID.
+        It might be better to do this using the session ID, but that involves
+        too many detailed changes to model manager calls.
+
+        May generate a ValueError if no GPU has been reserved.
+        """
         pass
 
     @property
@@ -105,16 +114,6 @@ class ModelCacheBase(ABC, Generic[T]):
         """Return true if the cache is configured to lazily offload models in VRAM."""
         pass
 
-    @abstractmethod
-    def offload_unlocked_models(self, size_required: int) -> None:
-        """Offload from VRAM any models not actively in use."""
-        pass
-
-    @abstractmethod
-    def move_model_to_device(self, cache_entry: CacheRecord[AnyModel], target_device: torch.device) -> None:
-        """Move model into the indicated device."""
-        pass
-
     @property
     @abstractmethod
     def stats(self) -> Optional[CacheStats]:
@@ -143,7 +142,6 @@ class ModelCacheBase(ABC, Generic[T]):
         self,
         key: str,
         model: T,
-        size: int,
         submodel_type: Optional[SubModelType] = None,
     ) -> None:
         """Store model under key and optional submodel_type."""
@@ -177,6 +175,11 @@ class ModelCacheBase(ABC, Generic[T]):
         """Return true if the model identified by key and submodel_type is in the cache."""
         pass
 
+    @abstractmethod
+    def model_to_device(self, cache_entry: CacheRecord[AnyModel], target_device: torch.device) -> AnyModel:
+        """Move a copy of the model into the indicated device and return it."""
+        pass
+
     @abstractmethod
     def cache_size(self) -> int:
         """Get the total size of the models currently cached."""