math benchmark
HMMT25 Leaderboard
HMMT25 is the Harvard-MIT Mathematics Tournament 2025, a student-organized mathematics competition for high school students with events at MIT in November 2025 and Harvard in February 2026.
Updated Sep 5, 2026
Models28
Model coverage28
MetricScore
EvidenceB
HMMT25 Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelXA | Score96.70% | Percentile100.00% | Participants28 | EvidenceC | Evaluated |
| Rank02 | ModelAC | Score94.60% | Percentile96.30% | Participants28 | EvidenceC | Evaluated |
| Rank03 | ModelAC | Score92.70% | Percentile92.59% | Participants28 | EvidenceC | Evaluated |
| Rank04 | ModelAC | Score90.70% | Percentile88.89% | Participants28 | EvidenceC | Evaluated |
| Rank05 | ModelAC | Score90.30% | Percentile85.19% | Participants28 | EvidenceC | Evaluated |
| Rank06 | ModelXA | Score90.00% | Percentile81.48% | Participants28 | EvidenceC | Evaluated |
| Rank07 | ModelAC | Score89.80% | Percentile77.78% | Participants28 | EvidenceC | Evaluated |
| Rank08 | ModelAC | Score89.20% | Percentile74.07% | Participants28 | EvidenceC | Evaluated |
| Rank09 | ModelIB | Score89.17% | Percentile70.37% | Participants28 | EvidenceC | Evaluated |
| Rank10 | ModelAC | Score89.10% | Percentile66.67% | Participants28 | EvidenceC | Evaluated |
| Rank11 | ModelSA | Score85.80% | Percentile62.96% | Participants28 | EvidenceC | Evaluated |
| Rank12 | ModelAC | Score83.90% | Percentile59.26% | Participants28 | EvidenceC | Evaluated |
| Rank13 | ModelAC | Score82.90% | Percentile55.56% | Participants28 | EvidenceC | Evaluated |
| Rank14 | ModelIB | Score78.33% | Percentile51.85% | Participants28 | EvidenceC | Evaluated |
| Rank15 | ModelAC | Score77.40% | Percentile48.15% | Participants28 | EvidenceC | Evaluated |
| Rank16 | ModelAC | Score76.80% | Percentile44.44% | Participants28 | EvidenceC | Evaluated |
| Rank17 | ModelSA | Score74.20% | Percentile40.74% | Participants28 | EvidenceC | Evaluated |
| Rank18 | ModelAC | Score73.90% | Percentile37.04% | Participants28 | EvidenceC | Evaluated |
| Rank19 | ModelAC | Score67.60% | Percentile33.33% | Participants28 | EvidenceC | Evaluated |
| Rank20 | ModelIB | Score66.67% | Percentile29.63% | Participants28 | EvidenceC | Evaluated |
| Rank21 | ModelAC | Score60.60% | Percentile25.93% | Participants28 | EvidenceC | Evaluated |
| Rank22 | ModelAC | Score57.40% | Percentile22.22% | Participants28 | EvidenceC | Evaluated |
| Rank23 | ModelAC | Score55.40% | Percentile18.52% | Participants28 | EvidenceC | Evaluated |
| Rank24 | ModelAC | Score54.10% | Percentile14.81% | Participants28 | EvidenceC | Evaluated |
| Rank25 | ModelAC | Score53.10% | Percentile11.11% | Participants28 | EvidenceC | Evaluated |
| Rank26 | ModelAC | Score50.60% | Percentile7.41% | Participants28 | EvidenceC | Evaluated |
| Rank27 | ModelAC | Score32.50% | Percentile3.70% | Participants28 | EvidenceC | Evaluated |
| Rank28 | ModelAC | Score30.70% | Percentile0.00% | Participants28 | 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.
What is HMMT25?
What HMMT25 measures and how its scores work.
HMMT25 is a mathematics competition featuring individual tests, team rounds, and guts rounds.
HMMT25 measures performance using the Score metric, reported as a ratio.
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
- No
- Evaluation key
- overall
LLM Stats. 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 96.70%.
What are the top three models on HMMT25?
The current leaders are Grok-4 Heavy (96.70%), Qwen3.6 Plus (94.60%), and Qwen3.5-397B-A17B (92.70%).
Which HMMT25 model has the lowest official input price?
Qwen3.6-35B-A3B has the lowest matched official input price at $0.25 input / $1.5 output per 1M tokens.
Which models are fastest among HMMT25 results?
The fastest matched records are Grok-4 (100.00 tok/s via xAI) and Qwen3 VL 4B Thinking (8.77 tok/s via DeepInfra).
Does the highest score result prove overall model quality?
No. This benchmark measures one defined capability or task. The overall LLMBoard score uses a separate aggregation across eligible benchmark evidence.