math benchmark
MATH-500 Leaderboard
MATH-500 is a benchmark subset of 500 challenging competition mathematics problems from AMC 10, AMC 12, AIME, and other mathematics competitions.
Updated Sep 5, 2026
Models32
Model coverage32
MetricScore
EvidenceC
MATH-500 Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelME | Score99.20% | Percentile100.00% | Participants32 | EvidenceC | Evaluated |
| Rank02 | ModelSA | Score98.60% | Percentile96.77% | Participants32 | EvidenceC | Evaluated |
| Rank03 | ModelZA | Score98.20% | Percentile93.55% | Participants32 | EvidenceC | Evaluated |
| Rank04 | ModelZA | Score98.10% | Percentile90.32% | Participants32 | EvidenceC | Evaluated |
| Rank05 | ModelNV | Score97.80% | Percentile87.10% | Participants32 | EvidenceC | Evaluated |
| Rank06 | ModelMA | Score97.40% | Percentile83.87% | Participants32 | EvidenceC | Evaluated |
| Rank07 | ModelMA | Score97.40% | Percentile80.65% | Participants32 | EvidenceC | Evaluated |
| Rank08 | ModelNV | Score97.00% | Percentile77.42% | Participants32 | EvidenceC | Evaluated |
| Rank09 | ModelSA | Score97.00% | Percentile74.19% | Participants32 | EvidenceC | Evaluated |
| Rank10 | ModelME | Score96.80% | Percentile70.97% | Participants32 | EvidenceC | Evaluated |
| Rank11 | ModelMI | Score96.80% | Percentile67.74% | Participants32 | EvidenceC | Evaluated |
| Rank12 | ModelNV | Score96.60% | Percentile64.52% | Participants32 | EvidenceC | Evaluated |
| Rank13 | ModelME | Score96.40% | Percentile61.29% | Participants32 | EvidenceC | Evaluated |
| Rank14 | ModelAN | Score96.20% | Percentile58.06% | Participants32 | EvidenceC | Evaluated |
| Rank15 | ModelMA | Score96.20% | Percentile54.84% | Participants32 | EvidenceC | Evaluated |
| Rank16 | ModelMI | Score96.00% | Percentile51.61% | Participants32 | EvidenceC | Evaluated |
| Rank17 | ModelDE | Score95.90% | Percentile48.39% | Participants32 | EvidenceC | Evaluated |
| Rank18 | ModelNV | Score95.40% | Percentile45.16% | Participants32 | EvidenceC | Evaluated |
| Rank19 | ModelMI | Score94.60% | Percentile41.94% | Participants32 | EvidenceC | Evaluated |
| Rank20 | ModelDE | Score94.50% | Percentile38.71% | Participants32 | EvidenceC | Evaluated |
| Rank21 | ModelDE | Score94.30% | Percentile35.48% | Participants32 | EvidenceC | Evaluated |
| Rank22 | ModelDE | Score94.00% | Percentile32.26% | Participants32 | EvidenceC | Evaluated |
| Rank23 | ModelDE | Score93.90% | Percentile29.03% | Participants32 | EvidenceC | Evaluated |
| Rank24 | ModelDE | Score92.80% | Percentile25.81% | Participants32 | EvidenceC | Evaluated |
| Rank25 | ModelAC | Score90.60% | Percentile22.58% | Participants32 | EvidenceC | Evaluated |
| Rank26 | ModelAC | Score90.60% | Percentile19.35% | Participants32 | EvidenceC | Evaluated |
| Rank27 | ModelDE | Score90.20% | Percentile16.13% | Participants32 | EvidenceC | Evaluated |
| Rank28 | ModelOP | Score90.00% | Percentile12.90% | Participants32 | EvidenceC | Evaluated |
| Rank29 | ModelDE | Score89.10% | Percentile9.68% | Participants32 | EvidenceC | Evaluated |
| Rank30 | ModelDE | Score83.90% | Percentile6.45% | Participants32 | EvidenceC | Evaluated |
MATH-500 Highlights
The leading models and scores on this benchmark.
MATH-500 Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for MATH-500
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is MATH-500?
What MATH-500 measures and how its scores work.
MATH-500 is a subset of the MATH dataset containing 500 problems with full step-by-step solutions across seven mathematical subjects and multiple difficulty levels.
MATH-500 reports a Score metric in ratio units.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of C.
- Family
- MATH-500
- 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 MATH-500.
Which model scores highest on MATH-500?
LongCat-Flash-Thinking is currently ranked first with 99.20%.
What are the top three models on MATH-500?
The current leaders are LongCat-Flash-Thinking (99.20%), Sarvam-105B (98.60%), and GLM-4.5 (98.20%).
Which MATH-500 model has the lowest official input price?
Phi 4 Mini Reasoning has the lowest matched official input price at $0.08 input / $0.30 output per 1M tokens.
Which models are fastest among MATH-500 results?
The fastest matched records are o1-mini (115.00 tok/s via OpenAI), DeepSeek-V3 0324 (100.00 tok/s via DeepSeek), and DeepSeek-V3 (100.00 tok/s via DeepSeek).
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.