← Back to resources

Benchmark

FrontierMath

Profile in progress

Epoch AI

Assess advanced mathematical reasoning on FrontierMath problems.

Open official page ↗

Resource profile

Access
Access status unknown
Access note
Version, split, tools and current results must accompany comparison.
Inputs
Model answers under the benchmark's stated evaluation setup.
Outputs
Benchmark results for the reported version and conditions.
License
License not yet verified
Limits
A benchmark result does not establish general mathematical capability outside its version, split and tool setting.
Evaluation conditions
Record benchmark version, split, model version, tools, budget and result source for every comparison.
Code release
Not verified
Data release
Not verified
Site execution check
Not verified

Associated timeline