# Harvard-MIT Mathematics Tournament February 2026 (HMMT Feb 2026)

> A February 2026 HMMT slice used in newer frontier-model math comparisons.

Canonical page: https://benchlm.ai/benchmarks/hmmtfeb2026

- Category: [Mathematics](/math)
- Last updated: September 15, 2026

## About HMMT Feb 2026

- Year: 2026
- Tasks: Competition math problems
- Format: Contest mathematics
- Difficulty: Olympiad-style mathematics
- Paper: [Qwen3.6 launch benchmarks](https://qwen.ai/blog?id=qwen3.6)

HMMT February 2026 matters because small score deltas at the frontier often depend on which contest set is used. BenchLM keeps this newer slice distinct from older HMMT summary rows.

HMMT Feb 2026 is currently weighted in BenchLM's scoring formula. The Mathematics category carries 5% of the overall score, and HMMT Feb 2026 contributes 25% of that category score.

## Leaderboard (23 models)

| Rank | Model | Creator | Score |
|------|-------|---------|-------|
| 1 | [Qwen3.7 Max](/models/qwen3-7-max) | Alibaba | 97.1% |
| 2 | [DeepSeek V4 Pro 0813](/models/deepseek-v4-pro-0813) | DeepSeek | 95.2% |
| 3 | [DeepSeek V4 Flash 0731](/models/deepseek-v4-flash-0731) | DeepSeek | 94.8% |
| 4 | [Solar Open 2](/models/solar-open-2) | Upstage | 93.9% |
| 5 | [Qwen3.7 Plus](/models/qwen3-7-plus) | Alibaba | 92.9% |
| 6 | [Kimi K2.6](/models/kimi-2-6) | Moonshot AI | 92.7% |
| 7 | [GLM-5.2](/models/glm-5-2) | Z.AI | 92.5% |
| 8 | [Inkling-Small](/models/inkling-small) | Thinking Machines Lab | 90.2% |
| 9 | [Qwen3.5 397B](/models/qwen3-5-397b) | Alibaba | 87.9% |
| 10 | [Qwen3.6 Plus](/models/qwen3-6-plus) | Alibaba | 87.8% |
| 11 | [Kimi K2.5](/models/kimi-k2-5) | Moonshot AI | 87.1% |
| 12 | [Ling 3.0 Flash](/models/ling-3-0-flash) | InclusionAI | 87.0% |
| 13 | [GLM-5](/models/glm-5) | Z.AI | 86.4% |
| 14 | [Claude Opus 4.5](/models/claude-opus-4-5) | Anthropic | 85.3% |
| 15 | [MAI-Thinking-1](/models/mai-thinking-1) | Microsoft | 84.9% |
| 16 | [Qwen3.6-27B](/models/qwen3-6-27b) | Alibaba | 84.3% |
| 17 | [Qwen3.6-35B-A3B](/models/qwen3-6-35b-a3b) | Alibaba | 83.6% |
| 18 | [GLM-5.1](/models/glm-5-1) | Z.AI | 82.6% |
| 19 | [K-EXAONE 2.0](/models/k-exaone-2-0) | LG AI Research | 78.4% |
| 20 | [ZAYA1-8B](/models/zaya1-8b) | Zyphra | 71.6% |
| 21 | [MiniCPM5-2B](/models/minicpm5-2b) | OpenBMB | 63.8% |
| 22 | [LongCat-Flash-Lite-Sparse](/models/longcat-flash-lite-sparse) | Meituan | 40.5% |
| 23 | [MiniCPM5-1B](/models/minicpm5-1b) | OpenBMB | 25.8% |

## FAQ

### What does HMMT Feb 2026 measure?

A February 2026 HMMT slice used in newer frontier-model math comparisons.

### Which model scores highest on HMMT Feb 2026?

Qwen3.7 Max by Alibaba currently leads with a score of 97.1% on HMMT Feb 2026.

### How many models are evaluated on HMMT Feb 2026?

23 AI models have been evaluated on HMMT Feb 2026 on BenchLM.

### Does HMMT Feb 2026 affect BenchLM's overall score?

Yes. HMMT Feb 2026 is a weighted benchmark inside the Mathematics category, which carries 5% of BenchLM's overall score. HMMT Feb 2026 itself contributes 25% of that category score.

## Compare Top Models on HMMT Feb 2026

- [Qwen3.7 Max vs DeepSeek V4 Pro 0813](/compare/deepseek-v4-pro-0813-vs-qwen3-7-max)
- [DeepSeek V4 Pro 0813 vs DeepSeek V4 Flash 0731](/compare/deepseek-v4-flash-0731-vs-deepseek-v4-pro-0813)
- [DeepSeek V4 Flash 0731 vs Solar Open 2](/compare/deepseek-v4-flash-0731-vs-solar-open-2)
- [Solar Open 2 vs Qwen3.7 Plus](/compare/qwen3-7-plus-vs-solar-open-2)