NexusRaven accepts a list of python functions.

The model can answer questions accurately:

“What is 1+10?”

“I have a cake that is about 3 centimenters high and 200 centimeters in diameter. How much cake do I have?”

Function calls have been improved according to the author. Still, it can be prone to generate incorrect calls.

Call function by LLM prompt

NexusRaven was trained on commercially viable data and is licensed under the Llama 2 community license following the original CodeLlama-13b-hf model.


Related Tool Assisted LLM:

Kani

Now if we only had a way to use SQLCoder and Raven together in a Multi Agent Chat.

    • noneabove1182@sh.itjust.worksM
      link
      fedilink
      English
      arrow-up
      1
      ·
      10 months ago

      Didn’t do well at rag, but maybe that’s because RAG is mostly handled by the wrapper rather than relying on the output of the model, think it’ll matter more for langchain/guidance and the example they gave