Commit Graph

52 Commits

Author SHA1 Message Date
psychedelicious
23d59abbd7 chore: ruff 2025-08-12 10:51:05 +10:00
psychedelicious
cf7fa5bce8 perf(backend): clear torch cache after encoding each image in kontext extension
Slightly reduces VRAM allocations.
2025-08-12 10:51:05 +10:00
psychedelicious
c6eff71b74 fix(backend): bug in kontext canvas dimension tracking when concating in latent space
We weren't tracking the canvas dimensions properly which coudl result in
FLUX not "seeing" ref images after the first very well
2025-08-12 10:51:05 +10:00
psychedelicious
7d86f00d82 feat(mm): implement working memory estimation for VAE encode for all models
Tell the model manager that we need some extra working memory for VAE
encoding operations to prevent OOMs.

See previous commit for investigation and determination of the magic
numbers used.

This safety measure is especially relevant now that we have FLUX Kontext
and may be encoding rather large ref images. Without the working memory
estimation we can OOM as we prepare for denoising.

See #8405 for an example of this issue on a very low VRAM system. It's
possible we can have the same issue on any GPU, though - just a matter
of hitting the right combination of models loaded.
2025-08-12 10:51:05 +10:00
psychedelicious
e522de33f8 refactor(nodes): roll back latent-space resizing of kontext images 2025-08-04 23:03:12 +10:00
psychedelicious
56d75e1c77 feat(backend): use VAE mean encoding for Kontext reference images
Use distribution mean without sampling noise for more stable and
consistent reference image encoding, matching ComfyUI implementation
2025-08-04 23:03:12 +10:00
psychedelicious
df77a12efe refactor(backend): use torchvision transforms for Kontext image preprocessing
Replace numpy-based normalization with torchvision transforms for
consistency with other image processing in the codebase
2025-08-04 23:03:12 +10:00
psychedelicious
faf662d12e refactor(backend): use BICUBIC resampling for Kontext images
Switch from LANCZOS to BICUBIC for smoother image resizing to reduce
artifacts in reference image processing
2025-08-04 23:03:12 +10:00
psychedelicious
44a7dfd486 fix(backend): use consistent idx_offset=1 for all Kontext images
Changes from per-image index offsets to a consistent value of 1 for
all reference images, matching the ComfyUI implementation
2025-08-04 23:03:12 +10:00
psychedelicious
bb15e5cf06 feat(backend): add spatial tiling for multiple Kontext reference images
Implements intelligent spatial tiling that arranges multiple reference
images in a virtual canvas, choosing between horizontal and vertical
placement to maintain a square-like aspect ratio
2025-08-04 23:03:12 +10:00
psychedelicious
1ed1fefa60 feat(nodes): support multiple kontext ref images
Images are concatenated in latent space.
2025-07-30 12:26:48 +10:00
psychedelicious
68571ece8f tidy(app): remove unused methods 2025-07-04 20:35:29 +10:00
psychedelicious
4887424ca3 chore: ruff 2025-07-04 20:35:29 +10:00
Kent Keirsey
28f6a20e71 format import block 2025-07-04 20:35:29 +10:00
Kent Keirsey
c4142e75b2 fix import 2025-07-04 20:35:29 +10:00
Kent Keirsey
fefe563127 fix resizing and versioning 2025-07-04 20:35:29 +10:00
Kent Keirsey
983cb5ebd2 ruff ruff 2025-07-04 19:24:44 +10:00
Kent Keirsey
52dbdb7118 ruff 2025-07-04 19:24:44 +10:00
Kent Keirsey
71e6f00e10 test fixes
fix

test

fix 2

fix 3

fix 4

yet another

attempt new fix

pray

more pray

lol
2025-07-04 19:24:44 +10:00
Kent Keirsey
51e1c56636 ruff 2025-06-27 18:27:46 +10:00
Kent Keirsey
ca1df60e54 Explain the Magic 2025-06-27 18:27:46 +10:00
Cursor Agent
7549c1250d Add FLUX Kontext conditioning support for reference images
Co-authored-by: kent <kent@invoke.ai>

Fix Kontext sequence length handling in Flux denoise invocation

Co-authored-by: kent <kent@invoke.ai>

Fix Kontext step callback to handle combined token sequences

Co-authored-by: kent <kent@invoke.ai>

fix ruff

Fix Flux Kontext
2025-06-27 18:27:46 +10:00
Ryan Dick
7e894ffe83 Consolidate InpaintExtension implementations for SD3 and FLUX. 2025-04-10 10:50:13 +10:00
Ryan Dick
4c02ba908a Add support for FLUX Redux masks. 2025-03-06 10:31:17 +11:00
Ryan Dick
f1fde792ee Get FLUX Redux working: model loading and inference. 2025-03-06 10:31:17 +11:00
Ryan Dick
607d19f4dd We should not trust the value of since the model could be partially-loaded. 2025-01-07 19:22:31 -05:00
Ryan Dick
64364e7911 Short-circuit if there are no region masks in FLUX and don't apply attention masking. 2024-11-27 22:40:10 +00:00
Ryan Dick
6565cea039 Comment unused _prepare_unrestricted_attn_mask(...) for future reference. 2024-11-27 22:16:44 +00:00
Ryan Dick
e970185161 Tweak flux regional prompting attention scheme based on latest experimentation. 2024-11-27 22:13:07 +00:00
Ryan Dick
b54463d294 Allow regional prompting background regions to attend to themselves and to the entire txt embedding. 2024-11-26 17:57:31 +00:00
Ryan Dick
faee79dc95 Distinguish between restricted and unrestricted attn masks in FLUX regional prompting. 2024-11-26 16:55:52 +00:00
Ryan Dick
e01f66b026 Apply regional attention masks in the single stream blocks in addition to the double stream blocks. 2024-11-25 22:40:08 +00:00
Ryan Dick
53abdde242 Update Flux RegionalPromptingExtension to prepare both a mask with restricted image self-attention and a mask with unrestricted image self attention. 2024-11-25 22:04:23 +00:00
Ryan Dick
94c088300f Be smarter about selecting the global CLIP embedding for FLUX regional prompting. 2024-11-25 20:15:04 +00:00
Ryan Dick
3741a6f5e0 Fix device handling for regional masks and apply the attention mask in the FLUX double stream block. 2024-11-25 16:02:03 +00:00
Ryan Dick
2c23b8414c Use a single global CLIP embedding for FLUX regional guidance. 2024-11-22 23:01:43 +00:00
Ryan Dick
20356c0746 Fixup the logic for preparing FLUX regional prompt attention masks. 2024-11-21 22:46:25 +00:00
Ryan Dick
bad1149504 WIP - add rough logic for preparing the FLUX regional prompting attention mask. 2024-11-20 22:29:36 +00:00
Ryan Dick
fda7aaa7ca Pass RegionalPromptingExtension down to the CustomDoubleStreamBlockProcessor in FLUX. 2024-11-20 19:48:04 +00:00
Ryan Dick
fdccdd52d5 Fixes to get XLabsIpAdapterExtension running. 2024-10-21 15:43:00 +00:00
Ryan Dick
31ffd73423 Initial draft of integrating FLUX IP-Adapter inference support. 2024-10-21 15:42:56 +00:00
Ryan Dick
4aace24f1f Reduce peak memory utilization when preparing FLUX controlnet inputs. 2024-10-10 00:18:46 +00:00
Ryan Dick
3953e60a4f Remove instantx_control_mode from FLUX ControlNet node. 2024-10-09 22:00:54 +00:00
Ryan Dick
8b1ef4b902 Fix bug with InstantX input image range. 2024-10-09 19:38:30 +00:00
Ryan Dick
859944f848 Fix support for InstantX non-union models (with no single blocks). 2024-10-09 18:51:53 +00:00
Ryan Dick
c78eeb1645 Shift the controlnet-type-specific logic into the specific ControlNet extensions and make the FLUX model controlnet-type-agnostic. 2024-10-09 16:12:09 +00:00
Ryan Dick
cd88723a80 Add instantx_control_mode param to FLUX ControlNet invocation. 2024-10-09 14:17:42 +00:00
Ryan Dick
dea6cbd599 Create a dedicated FLUX ControlNet invocation. 2024-10-09 14:17:42 +00:00
Ryan Dick
0dd9f1f772 Bugfixes to get InstantX ControlNet working. 2024-10-09 14:17:42 +00:00
Ryan Dick
a783539cd2 Fix circular imports related to XLabsControlNetFluxOutput and InstantXControlNetFluxOutput. 2024-10-09 14:17:42 +00:00