Benchmark
HMMT February 2026
Math benchmarks measure how well models solve maths problems. Scores are percentages.
All math benchmarksScores from OpenEvals Leaderboard data.
- Models measured
- 5
- Scores recorded
- 5
- Sources
- 1
- IQ weight
- 0.41
Score distribution
- 86.4%
- Mean
- 86.4%
- Median
- 1.3%
- Std dev
- 84.1%
- 10th percentile
- 87.9%
- 90th percentile
- 84.1%
- Min
This benchmark has a weight of 0.41 for chat models, measured over 5 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | Qwen3.5 397B A17B· OpenEvals Leaderboard data· Reported by the model's developer. The source relaying it did not verify it. | 87.9% |
| 2 | Kimi K2.5· OpenEvals Leaderboard data· Reported by the model's developer. The source relaying it did not verify it. | 87.1% |
| 3 | GLM 5· OpenEvals Leaderboard data· Reported by the model's developer. The source relaying it did not verify it. | 86.4% |
| 4 | Step 3.5 Flash· OpenEvals Leaderboard data· Reported by the model's developer. The source relaying it did not verify it. | 86.4% |
| 5 | DeepSeek V3.2· OpenEvals Leaderboard data· Reported by the model's developer. The source relaying it did not verify it. | 84.1% |
Data credits: OpenEvals/leaderboard-data, MIT (licence)