retrieval-api

Retrieval API that utilizes llama-index

Installation

pip install -r requirements.txt

Startup

uvicorn main:app --reload

Access

http://localhost:8000/docs

Loaders

PDF DOCX IPYNB

TODO

• Add API endpoint for chatting with content (chat history etc.) ref. https://gpt-index.readthedocs.io/en/stable/core_modules/query_modules/chat_engines/usage_pattern.html
• Add more loaders
• Create dynamic loader for files (one endpoint)