TinyLLM

• Fixed bug with baseprompt updates to respond to saved Settings or new sessions.
• Updated baseprompt to include date and guidance for complex and open-ended questions.
• Add TZ local timezone environmental variable to ensure correct date in baseprompt.

Full Changelog: https://github.com/jasonacox/TinyLLM/compare/v0.14.0...v0.14.1

• Added ability to change LLM Temperature and MaxTokens in settings.
• Added optional prompt settings read-only options to allow viewing but prevent changes (PROMPT_RO=true).

Full Changelog: https://github.com/jasonacox/TinyLLM/compare/v0.13.0...v0.14.0

What's Changed

• Moved from Qdrant to Weaviate - This externalizes the sentence transformation work and lets the chatbot run as a smaller service. Activate by setting WEAVIATE_HOST to the address of the DB
• Added "References" text to output from /rag queries.
• Added ONESHOT environmental variable that if True will remove conversation threading allowing each query to be answered as a standalone sessions.
• Added RAG_ONLY environmental variable that if True will assume all queries should be directed to the default RAG database as set by WEAVIATE_LIBRARY.
• See https://github.com/jasonacox/TinyLLM/pull/5

docker run \
    -d \
    -p 5000:5000 \
    -e PORT=5000 \
    -e OPENAI_API_BASE="http://localhost:8000/v1" \
    -e ONESHOT="true" \
    -e RAG_ONLY="false" \
    -e WEAVIATE_HOST="localhost" \
    -e WEAVIATE_LIBRARY="tinyllm" \
    -v $PWD/.tinyllm:/app/.tinyllm \
    --name chatbot \
    --restart unless-stopped \
    jasonacox/chatbot

Full Changelog: https://github.com/jasonacox/TinyLLM/compare/v0.12.5...v0.13.0

• Added logic to poll LLM for model list. If only one model is available, use that. Otherwise verify the user requested model is available.
• Chatbot UI now shows model name and adds responsive elements to better display on mobile devices.
• Add encoding user prompts to correctly display html code in Chatbot.
• Fix chat.py CLI chatbot to handle user/assistant prompts for vLLM.

• Add encoding user prompts to correctly display html code in Chatbot.
• Fix chat.py CLI chatbot to handle user/assistant prompts for vLLM.

Full Changelog: https://github.com/jasonacox/TinyLLM/compare/v0.12.3...v0.12.4

• Bug fix for handle_url_prompt() to extract text from URL.

Full Changelog: https://github.com/jasonacox/TinyLLM/compare/v0.12.2...v0.12.3

v0.12.2

• Speed up command functions using aiohttp.
• Fix prompt_expand for rag command.
• Added topic option to /news command.

v0.12.1 - Performance Improvements

• Speed up user prompt echo. Immediately send to chat windows instead of waiting for LLM stream to start.
• Optimize message handling dispatching using async.
• Use AsyncOpenAI for non-streamed queries.

Full Changelog: https://github.com/jasonacox/TinyLLM/compare/v0.12.0...v0.12.2

Chatbot - v0.12.0 - FastAPI and Uvicorn

• Ported Chatbot to the async FastAPI and Uvicorn ASGI high speed web server implementation (https://github.com/jasonacox/TinyLLM/issues/3).
• Added /stats page to display configuration settings and current stats (optional ?format=json)
• UI updated to help enforce focus on text entry box.
• Moved prompts.json and Sentence Transformer model location to a ./.tinyllm for Docker support.

• Add /stats URL to Chatbot for settings and current status information.
• Update Chatbot HTML to set focus on user textbox.
• Move prompts.json and Sentence Transformer models into .tinyllm directory.

Full Changelog: https://github.com/jasonacox/TinyLLM/compare/v0.11.3...v0.11.4

What's Changed

• Fix docker PWD and network by @the023 in https://github.com/jasonacox/TinyLLM/pull/2
• Improve Chatbot for Docker
• Added admin alert broadcast feature (POST /alert)

New Contributors

• @the023 made their first contribution in https://github.com/jasonacox/TinyLLM/pull/2

Full Changelog: https://github.com/jasonacox/TinyLLM/commits/v0.11.3