Skip to content

Conversation

@nikil-ravi
Copy link

This PR adds IFEval. So far,I've tested baseline evaluation with Qwen 3 1.7B so far on an H100, and this runs in just under 15 minutes.

@max-andr
Copy link
Member

thanks! we need to discuss internally whether we want to add IFEval to our "official" suite of benchmarks. but i personally like IFEval, since IFEval scores can be improved a lot by doing smarter post-training.

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