A minimal Retrieval-Augmented Generation app with Streamlit and Foundry Local.
-
Make sure Foundry Local is running on
http://localhost:8000 -
Copy
.env.exampleto.envand configure:
FOUNDRY_API_KEY=your-foundry-api-key
FOUNDRY_API_URL=http://localhost:8000
FOUNDRY_MODEL=llama2
FOUNDRY_EMBEDDING_MODEL=nomic-embed-text
- Install dependencies:
pip install -r requirements.txt- Run:
streamlit run app.py- Upload a document (PDF, DOCX, or TXT) and ask questions about it!
- 📄 Upload documents in multiple formats
- 🔍 Semantic search via Foundry embeddings
- 💬 Chat with local Foundry LLM
- 📖 View source citations
app.py- Main Streamlit applicationconfig.py- Foundry configurationdocument_manager.py- Document processingvector_store.py- Embeddings and search