Add community evaluation results for AIME 2026

#82
by nielsr HF Staff - opened

This PR adds community-provided evaluation results for the following benchmarks:

These results were extracted from Appendix B.3 of the technical report, at maximum reasoning effort (100). This is based on the new evaluation results feature.

Note: This is an automated PR. Please review the evaluation results before merging.

Ready to merge
This branch is ready to get merged automatically.

Sign up or log in to comment