math benchmark
HMMT Feb 26
HMMT February 2026 is a math competition benchmark based on problems from the Harvard-MIT Mathematics Tournament, testing advanced mathematical problem-solving and reasoning.
Updated Aug 7, 2026
Published models11
Registry coverage11
MetricScore
EvidenceB
HMMT Feb 26 leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | AC | 97.1% | 100.0% | 11 | C | |
| 2 | DE | 95.2% | 90.0% | 11 | C | |
| 3 | DE | 94.8% | 80.0% | 11 | C | |
| 4 | AC | 92.9% | 70.0% | 11 | C | |
| 5 | MA | 92.7% | 60.0% | 11 | C | |
| 6 | ZA | 92.5% | 50.0% | 11 | C | |
| 7 | AC | 87.8% | 40.0% | 11 | C | |
| 8 | MI | 84.9% | 30.0% | 11 | C | |
| 9 | AC | 84.3% | 20.0% | 11 | C | |
| 10 | AC | 83.6% | 10.0% | 11 | C | |
| 11 | ZA | 82.6% | 0.0% | 11 | C |
Score distribution
Top published rows on the benchmark's original scale.
HMMT Feb 26 highlights
The top published results on this benchmark's own scale.
What is HMMT Feb 26?
Definition and scoring fields from the benchmark registry.
HMMT February 2026 is a math competition benchmark based on problems from the Harvard-MIT Mathematics Tournament, testing advanced mathematical problem-solving and reasoning.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- HMMT Feb 26
- Modality
- text
- Primary category
- math
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- hmmt-feb-26|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about HMMT Feb 26.
Which model scores highest on HMMT Feb 26?
Qwen3.7 Max is currently ranked first with 97.1%.
What does HMMT Feb 26 measure?
HMMT February 2026 is a math competition benchmark based on problems from the Harvard-MIT Mathematics Tournament, testing advanced mathematical problem-solving and reasoning.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
11 unique published model results are currently shown.
Does this benchmark affect the overall score?
This benchmark is marked as eligible for the current LLMBoard capability methodology.