Skip to content
This repository was archived by the owner on Jan 2, 2025. It is now read-only.

Conversation

@rmuller-ml
Copy link
Contributor

  • We separate the model from agent and answer to have different headrooms
  • We don't use the full 128k context window (only 24k)

@rmuller-ml rmuller-ml marked this pull request as ready for review November 30, 2023 18:54
Comment on lines +60 to +61
pub answer_model: model::LLMModel,
pub agent_model: model::LLMModel,
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

i think its confusing that Agent has, within itself, an agent_model field, can we call this something else? such as decsicion_model or reasoning_model?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think agent_model is fine.
The abstraction is: the user queries to get an answer and we have an answer_model used in the tool answer, to get the the correct context to answer the query we start an agent which has its own agent_model

Third opinion @ggordonhall ?

@ggordonhall ggordonhall merged commit 0fb9b62 into main Dec 7, 2023
@ggordonhall ggordonhall deleted the gpt4turbo_agent_separation branch December 7, 2023 18:13
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants