math benchmark
HMMT25 Leaderboard
Harvard-MIT Mathematics Tournament 2025 - A prestigious student-organized mathematics competition for high school students featuring two tournaments (November 2025 at MIT and February 2026 at Harvard) with individual tests, team rounds, and guts rounds
Updated Aug 17, 2026
Models25
Model coverage25
MetricScore
EvidenceB
HMMT25 Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelXA | Score96.7% | Percentile100.0% | Participants25 | EvidenceC | Evaluated |
| Rank02 | ModelAC | Score94.6% | Percentile95.8% | Participants25 | EvidenceC | Evaluated |
| Rank03 | ModelAC | Score92.7% | Percentile91.7% | Participants25 | EvidenceC | Evaluated |
| Rank04 | ModelAC | Score90.7% | Percentile87.5% | Participants25 | EvidenceC | Evaluated |
| Rank05 | ModelAC | Score90.3% | Percentile83.3% | Participants25 | EvidenceC | Evaluated |
| Rank06 | ModelXA | Score90.0% | Percentile79.2% | Participants25 | EvidenceC | Evaluated |
| Rank07 | ModelAC | Score89.8% | Percentile75.0% | Participants25 | EvidenceC | Evaluated |
| Rank08 | ModelAC | Score89.2% | Percentile70.8% | Participants25 | EvidenceC | Evaluated |
| Rank09 | ModelAC | Score89.1% | Percentile66.7% | Participants25 | EvidenceC | Evaluated |
| Rank10 | ModelSA | Score85.8% | Percentile62.5% | Participants25 | EvidenceC | Evaluated |
| Rank11 | ModelAC | Score83.9% | Percentile58.3% | Participants25 | EvidenceC | Evaluated |
| Rank12 | ModelAC | Score82.9% | Percentile54.2% | Participants25 | EvidenceC | Evaluated |
| Rank13 | ModelAC | Score77.4% | Percentile50.0% | Participants25 | EvidenceC | Evaluated |
| Rank14 | ModelAC | Score76.8% | Percentile45.8% | Participants25 | EvidenceC | Evaluated |
| Rank15 | ModelSA | Score74.2% | Percentile41.7% | Participants25 | EvidenceC | Evaluated |
| Rank16 | ModelAC | Score73.9% | Percentile37.5% | Participants25 | EvidenceC | Evaluated |
| Rank17 | ModelAC | Score67.6% | Percentile33.3% | Participants25 | EvidenceC | Evaluated |
| Rank18 | ModelAC | Score60.6% | Percentile29.2% | Participants25 | EvidenceC | Evaluated |
| Rank19 | ModelAC | Score57.4% | Percentile25.0% | Participants25 | EvidenceC | Evaluated |
| Rank20 | ModelAC | Score55.4% | Percentile20.8% | Participants25 | EvidenceC | Evaluated |
| Rank21 | ModelAC | Score54.1% | Percentile16.7% | Participants25 | EvidenceC | Evaluated |
| Rank22 | ModelAC | Score53.1% | Percentile12.5% | Participants25 | EvidenceC | Evaluated |
| Rank23 | ModelAC | Score50.6% | Percentile8.3% | Participants25 | EvidenceC | Evaluated |
| Rank24 | ModelAC | Score32.5% | Percentile4.2% | Participants25 | EvidenceC | Evaluated |
| Rank25 | ModelAC | Score30.7% | Percentile0.0% | Participants25 | EvidenceC | Evaluated |
HMMT25 Highlights
The leading models and scores on this benchmark.
HMMT25 Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for HMMT25
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
Ranking basisThis hmmt25 AI model leaderboard uses descending score in the benchmark's original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.
- 01XAGrok-4 HeavyxAI
- Score
- 96.7%
Strengths
- Ranks #1 of 25 compared models
- 100th percentile on this benchmark
- C evidence result
Considerations
- This result measures HMMT25, not total model capability
- 02ACQwen3.6 PlusAlibaba Cloud / Qwen Team
- Score
- 94.6%
- Price
- $0.50 input / $3.0 output per 1M tokens
- Speed
- Up to 16 tok/s via Together
Strengths
- Ranks #2 of 25 compared models
- 96th percentile on this benchmark
- C evidence result
Considerations
- This result measures HMMT25, not total model capability
- 03ACQwen3.5-397B-A17BAlibaba Cloud / Qwen Team
- Score
- 92.7%
- Price
- $0.60 input / $3.6 output per 1M tokens
Strengths
- Ranks #3 of 25 compared models
- 92th percentile on this benchmark
- C evidence result
Considerations
- This result measures HMMT25, not total model capability
- 04ACQwen3.6-27BAlibaba Cloud / Qwen Team
- Score
- 90.7%
- Price
- $0.60 input / $3.6 output per 1M tokens
- Speed
- Up to 6.1 tok/s via Novita
Strengths
- Ranks #4 of 25 compared models
- 88th percentile on this benchmark
- C evidence result
Considerations
- This result measures HMMT25, not total model capability
- 05ACQwen3.5-122B-A10BAlibaba Cloud / Qwen Team
- Score
- 90.3%
- Price
- $0.40 input / $3.2 output per 1M tokens
Strengths
- Ranks #5 of 25 compared models
- 83th percentile on this benchmark
- C evidence result
Considerations
- This result measures HMMT25, not total model capability
Selection summary
Best AI Models for HMMT25
Grok-4 Heavy currently leads HMMT25 with 96.7%. It is the top model on this specific benchmark, while the best LLM for the broader task should also be checked against other benchmarks, price and runtime.
Use this leaderboard with the supporting benchmark results and coverage details above. A leaderboard position summarizes the selected ranking signal; it does not replace workload-specific testing.
What is HMMT25?
What HMMT25 measures and how its scores work.
Harvard-MIT Mathematics Tournament 2025 - A prestigious student-organized mathematics competition for high school students featuring two tournaments (November 2025 at MIT and February 2026 at Harvard) with individual tests, team rounds, and guts rounds
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- HMMT25
- Modality
- text
- Primary category
- math
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- hmmt25|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about HMMT25.
Which model scores highest on HMMT25?
Grok-4 Heavy is currently ranked first with 96.7%.
What does HMMT25 measure?
Harvard-MIT Mathematics Tournament 2025 - A prestigious student-organized mathematics competition for high school students featuring two tournaments (November 2025 at MIT and February 2026 at Harvard) with individual tests, team rounds, and guts rounds
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
25 model results are currently shown.
Does this benchmark affect the overall score?
Yes. This benchmark can contribute to the current LLMBoard capability score.