This is an MVP of a document analysis AI-based research assistant. Originally using AWS Bedrock + Claude3, then Ollama... Now using Google Gemini.
- Generate Google Gemini API Key
- Create config.ini and paste API key
- In Console, use command
streamlit run app.py
- Replace AWS Bedrock with Ollama Llama3.2 (or :vision) FM
- Replace Ollama with Google Gemini
- Set up LLM to ingest more than one image at a time
- Add feature to intake PDF's
- Refactor
- Add unit tests
- Beta test and add features