BeyondAIME
huggingface.co- Category
- Developer Tools
- Listed in
- #51 Find AI benchmarks
- Type
- TOOL
- Date
About
ByteDance Seed's test set of competition mathematics at or above the difficulty of the hardest AIME problems. Every problem is manually rewritten to resist contamination, needs no knowledge beyond university level, and has an integer answer that is as hard to guess as to derive.
Why it made the leaderboard
Compare the task, benchmark version, harness, and grading method before using model scores to choose a model.
Tags
benchmarkevaluation
Comments (0)
No comments yet
Editorially curated, with community endorsements as a secondary signal. Corrections welcome.