txtai claims to be an all-in-one embeddings database for semantic search, LLM orchestration and language model workflows. From what I’ve seen this could potentially be a Langchain alternative when it comes to RAG, much like how llama-index is better at certain aspects. The trade-off here is that there’s less glue holding everything together and there is a noticeable paradigm shift to get used.

This page will be a collection of useful resource I’ve found while getting familiar with this library.

Use External Embeddings

Create an Index

Persistent Index

Use External LLM

RAG