Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add Cohere Command as an LLM backend #127

Open
wants to merge 2 commits into
base: main
Choose a base branch
from

Conversation

malexw
Copy link

@malexw malexw commented Jan 5, 2024

This PR demonstrates adding Cohere's Command as a backend LLM for OpenGPTs. The code uses the XML agent as a starting point and makes some adjustments to the prompt, as well as adding a new method to stringify the tool schema in a way that Command responds well to.

One known issue is that even though Command does seem to do a good job of calling functions when appropriate, it still includes a normal text response after the function call. I'll trim that out from the log in the future, but I've left it in for now for testing and debugging.

@malexw malexw mentioned this pull request Jan 5, 2024
@malexw malexw changed the title Initial testing of Command as an LLM backend for OpenGPTs Add Cohere Command as an LLM backend Jan 5, 2024
@malexw malexw marked this pull request as ready for review January 8, 2024 19:14
@giannis2two
Copy link

Thanks for opening up the PR @malexw.

@nfcampos do you think the PR implementation is an acceptable design? Have noticed in the repo that similar LLM additions were added under backend/packages/gizmo-agent/gizmo_agent/agent_types/xml/ vs PR Cohere support is added under backend/packages/gizmo-agent/gizmo_agent/agent_types/. Thanks in advance.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants