Back to Benchmarks
📐
HMMT
Harvard-MIT Math Tournament
Harvard-MIT math competition problems
📐 Math
Best: 97.7%
4 Models Evaluated
Champion
GPT-5.4 Pro
OpenAI
97.7%
Best Score
Model Rankings
All models ranked by score on HMMT. Scores are relative to the best score of 97.7%.
| # | Model | Provider | Score | vs Best |
|---|---|---|---|---|
| 1 | GPT-5.4 Pro Elo 1480 | OpenAI | 97.7% | Best |
| 2 | GPT-5.4 Elo 1463 | OpenAI | 97% | -0.7% |
| 3 | Gemini 3 Pro Elo 1486 | 96% | -1.7% | |
| 4 | DeepSeek V4 Pro (Max) | DeepSeek | 95.2% | -2.5% |
Score Distribution
GPT-5.4 Pro
97.7%
GPT-5.4
97%
Gemini 3 Pro
96%
DeepSeek V4 Pro (Max)
95.2%
Run This Benchmark
Learn how to evaluate your own models on HMMT.
HMMT 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.