Updated daily at 05:00 UTC
Benchmark
FrontierMath Erdős
Measures: 68 problems posed or studied by Paul Erdős that were open as of August 2026; the model must write a complete Lean 4 proof or disproof, checked by the Lean kernel in a sandbox; scored as the share of problems solved.
- Maintainer
- Epoch AI
- Introduced
- 2026
- Status
- Active
- Saturation
- Headroom
- Typical provenance
- Independent
- In scoring set
- Yes
Leaderboard
Independent and third-party results only.
No independent or third-party results yet. Show all results