Opiniated RAG for integrating GenAI in your apps 🧠 Focus on your product rather than the RAG. Easy integration in existing products with customisation! Any LLM: GPT4, Groq, Llama. Any Vectorstore: PGVector, Faiss. Any Files. Anyway you want.
Currently, document chunks are stored individually into our vector database (PGVector), i.e. the only relationship we record is the one between a chunk and its original document.
We should expand this to extract the document layout (headers, footers, table, image, caption, …) and the relationships (chunk --> page --> file, previous_chunk --> chunk --> next_chunk, …) and store them into a database, see our scheme.
linear[bot]
commented
3 weeks ago
Currently, document chunks are stored individually into our vector database (PGVector), i.e. the only relationship we record is the one between a chunk and its original document.
We should expand this to extract the document layout (headers, footers, table, image, caption, …) and the relationships (chunk --> page --> file, previous_chunk --> chunk --> next_chunk, …) and store them into a database, see our scheme.