search

Holmeswww / AgentKit

An intuitive LLM prompting framework for multifunctional agents, by explicitly constructing a complex "thought process" from simple natural language prompts.
Creative Commons Attribution 4.0 International
269 stars 26 forks source link

Beautiful work! #4

Closed aldopareja closed 2 months ago

aldopareja commented 2 months ago

Wondering how hard would it be to leverage langchain to create one node, that enables changing the model being called at different points plus all of the other tools developed there.

Holmeswww commented 2 months ago

Hi,

Each Node can take a custom compose prompt function. In our experiments, we actually mixed different LLM models for different nodes.

You can also use langchain inside the node by using a custom compose function and LLM-querying function.

For example, you can provide an LLM-query function that uses langchain instead of OpenAI.

I wonder if this addresses your question.