Skip to content

Comment on Show HN: LLMWare – Small Specialized Function Calling 1B LLMs for Multi-Step RAG

Comments

This looks very interesting as majority of the world is moving towards bigger "do everything for you" LLMs. Just few preliminary questions after I glanced over the repo and blogs:

- there are also many small models that try to "do everything" like phi, mistral, etc., do you find llmware have better quality and performance? - how does the rag relate features compare with other tools like llamaindex?

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.