This commit fixes an issue where Retrieval-Augmented Generation (RAG)
queries were still being generated even when all attached files were set
to 'full context' mode. This was inefficient as the full content of the
files was already available to the model.
The `chat_completion_files_handler` in `backend/open_webui/utils/middleware.py`
has been updated to:
- Check if all attached files have the `context: 'full'` property.
- Skip the `generate_queries` step if all files are in full context mode.
- Pass a `full_context=True` flag to the `get_sources_from_items`
function to ensure it fetches the entire document content instead of
performing a vector search.
This change ensures that RAG queries are only generated when necessary,
improving the efficiency of the system.
This commit fixes a bug where hidden models were still visible and selectable in the notes section. The following changes were made:
- The model dropdown in the chat section of the note editor is now filtered to exclude hidden models.
- The default model selection logic in the note editor is now filtered to exclude hidden models.
Previously, when downloading a generated image, the filename was set to the AI's response text. This was not ideal as the response text could be long and contain characters that are not suitable for filenames.
This commit changes the behavior to use a generic, translatable filename for downloaded images. The new filename is 'ai_generated_image' (or its translation), which is more user-friendly and consistent.
