Merge pull request #300 from acon96/release/v0.4.1

Release v0.4.1

README.md

@@ -158,6 +158,7 @@ python3 train.py \
 ## Version History
 | Version | Description                                                                                                                                                                                                          |
 |---------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
+| v0.4.1  | Fix an issue with using Llama.cpp models downloaded from HuggingFace                                                                                                                                                 |
 | v0.4    | Rewrite integration to support tool calling models/agentic tool use loop, voice streaming, multiple config sub-entries per backend, and dynamic llama.cpp processor selection                                        |
 | v0.3.11 | Bug-fixes and llama.cpp version update                                                                                                                                                                               |
 | v0.3.10 | Add support for the OpenAI "Responses" API endpoint, Update llama.cpp version, Fix for breaking change in HA version 2025.7.0                                                                                        |

custom_components/llama_conversation/__init__.py

@@ -2,6 +2,7 @@
 from __future__ import annotations
 
 import logging
+import os
 from typing import Final
 
 from homeassistant.config_entries import ConfigEntry, ConfigSubentry
@@ -50,7 +51,7 @@ from .backends.llamacpp import LlamaCppClient
 from .backends.generic_openai import GenericOpenAIAPIClient, GenericOpenAIResponsesAPIClient
 from .backends.tailored_openai import TextGenerationWebuiClient, LlamaCppServerClient
 from .backends.ollama import OllamaAPIClient
-from .utils import get_llama_cpp_python_version
+from .utils import get_llama_cpp_python_version, download_model_from_hf
 
 _LOGGER = logging.getLogger(__name__)
 
@@ -161,6 +162,27 @@ async def async_migrate_entry(hass: HomeAssistant, config_entry: LocalLLMConfigE
         hass.config_entries.async_update_entry(config_entry, title=entry_title, data=entry_data, options=entry_options, version=3)
 
         _LOGGER.debug("Migration to subentries complete")
+    
+    if config_entry.version == 3 and config_entry.minor_version == 0:
+        # add the downloaded model file to options if missing
+        if config_entry.data.get(CONF_BACKEND_TYPE) == BACKEND_TYPE_LLAMA_CPP:
+            for subentry in config_entry.subentries.values():
+                if subentry.data.get(CONF_DOWNLOADED_MODEL_FILE) is None:
+                    model_name = subentry.data[CONF_CHAT_MODEL]
+                    quantization_type = subentry.data[CONF_DOWNLOADED_MODEL_QUANTIZATION]
+                    storage_folder = os.path.join(hass.config.media_dirs.get("local", hass.config.path("media")), "models")
+
+                    new_options = dict(subentry.data)
+                    file_name = await hass.async_add_executor_job(download_model_from_hf, model_name, quantization_type, storage_folder, True)
+                    new_options[CONF_DOWNLOADED_MODEL_FILE] = file_name
+
+                    hass.config_entries.async_update_subentry(
+                        config_entry, subentry, data=MappingProxyType(new_options)
+                    )
+
+        hass.config_entries.async_update_entry(config_entry, minor_version=1)
+        
+        _LOGGER.debug("Migration to add downloaded model file complete") 
 
     return True
 

custom_components/llama_conversation/config_flow.py

@@ -220,7 +220,7 @@ class ConfigFlow(BaseConfigFlow, domain=DOMAIN):
     """Handle a config flow for Local LLM Conversation."""
 
     VERSION = 3
-    MINOR_VERSION = 0
+    MINOR_VERSION = 1
 
     install_wheel_task = None
     install_wheel_error = None
@@ -1071,11 +1071,10 @@ class LocalLLMSubentryFlowHandler(ConfigSubentryFlow):
         if not self.download_task:
             model_name = self.model_config[CONF_CHAT_MODEL]
             quantization_type = self.model_config[CONF_DOWNLOADED_MODEL_QUANTIZATION]
-
             storage_folder = os.path.join(self.hass.config.media_dirs.get("local", self.hass.config.path("media")), "models")
 
             async def download_task():
-                await self.hass.async_add_executor_job(
+                return await self.hass.async_add_executor_job(
                     download_model_from_hf, model_name, quantization_type, storage_folder
                 )
 

custom_components/llama_conversation/const.py

@@ -317,5 +317,5 @@ OPTIONS_OVERRIDES = {
     },
 }
 
-INTEGRATION_VERSION = "0.4.0"
+INTEGRATION_VERSION = "0.4.1"
 EMBEDDED_LLAMA_CPP_PYTHON_VERSION = "0.3.16+b6153"

custom_components/llama_conversation/manifest.json

@@ -1,7 +1,7 @@
 {
   "domain": "llama_conversation",
   "name": "Local LLMs",
-  "version": "0.4.0",
+  "version": "0.4.1",
   "codeowners": ["@acon96"],
   "config_flow": true,
   "dependencies": ["conversation"],

custom_components/llama_conversation/utils.py

@@ -136,7 +136,7 @@ def custom_custom_serializer(value):
     
     return cv.custom_serializer(value)
 
-def download_model_from_hf(model_name: str, quantization_type: str, storage_folder: str):
+def download_model_from_hf(model_name: str, quantization_type: str, storage_folder: str, file_lookup_only: bool = False):
     try:
         from huggingface_hub import hf_hub_download, HfFileSystem
     except Exception as ex:
@@ -162,6 +162,7 @@ def download_model_from_hf(model_name: str, quantization_type: str, storage_fold
         repo_type="model",
         filename=wanted_file[0].removeprefix(model_name + "/"),
         cache_dir=storage_folder,
+        local_files_only=file_lookup_only
     )
 
 def _load_extension():