Text-to-SQL using LLM and RAG
pip3 install -r requirements.txt
bin/run-local.sh
LLM:
- platform:
- OpenAI ChatGPT
- Cohere Command R+ ?
- Claude
- Meta Llama 3 ?
RAG:
- database:
- chromadb (standalone)
- postgres pgvector
- embedding model:
- sentence transformers:
- all-MiniLM-L6-v2 (*max len = 256)
- sentence transformers:
- index (VSS):
- HNSW
- IVF-PQ ?
- Implement standalone application with ChromaDB
- Implement CRUD with an admin page
- Set a docker environment
- Implement a vector store for PGVector