Back to Benchmarks
📐

FrontierMath

Frontier Mathematics

Cutting-edge math research problems

📐 Math
Best: 52.4%
2 Models Evaluated
Champion

GPT-5.4 Pro

OpenAI

51.7%

Best Score

Model Rankings

All models ranked by score on FrontierMath. Scores are relative to the best score of 52.4%.

#ModelProviderScorevs Best
1GPT-5.4 Pro
Elo 1480
OpenAI
51.7%
Best
2DeepSeek V4 Pro (Max)DeepSeek
38.3%
-14.1%

Score Distribution

GPT-5.4 Pro
51.7%
DeepSeek V4 Pro (Max)
38.3%

Run This Benchmark

Learn how to evaluate your own models on FrontierMath.

FrontierMath is a standardized benchmark for evaluating AI model capabilities.

To run this benchmark on your own models, visit the official benchmark page or leaderboard for detailed instructions, code, and evaluation scripts.

Data compiled from official model cards, public leaderboards, and independent evaluations.