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

Empty responses using o1-preview #522

Open
matthewgertner opened this issue Nov 14, 2024 · 2 comments
Open

Empty responses using o1-preview #522

matthewgertner opened this issue Nov 14, 2024 · 2 comments

Comments

@matthewgertner
Copy link

I selected o1-preview as the model and turned off streaming. For short prompts, it seems to be fine. But when I pasted in a large block of JSON, I just get empty responses with the cost of the message in the footer (e.g. "o1-preview-2024-09-12 using 1269 tokens ~= $0.042075").

Any idea what might be causing this?

@Ryan526
Copy link
Contributor

Ryan526 commented Nov 18, 2024

In your profile for o1 preview change token settings to fit the max o1 can handle. Reasoning tokens are using up everything so it spits out a blank response if you don't do this.

Max Context: 128k
Max Output 32k

@matthewgertner
Copy link
Author

I could only see one setting for max tokens. When I reverted it to the default of 128k, it started working as expected.

Thanks!

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

No branches or pull requests

2 participants