math benchmark
MathVista Leaderboard
MathVista evaluates mathematical reasoning of foundation models in visual contexts. It consists of 6,141 examples derived from 28 existing multimodal datasets and 3 newly created datasets (IQTest, FunctionQA, and PaperQA), combining challenges from diverse mathematical and visual tasks to assess models' ability to understand complex figures and perform rigorous reasoning.
Updated Aug 17, 2026
Models39
Model coverage39
MetricScore
EvidenceB
MathVista Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelBY | Score90.7% | Percentile100.0% | Participants39 | EvidenceC | Evaluated |
| Rank02 | ModelBY | Score90.5% | Percentile97.4% | Participants39 | EvidenceC | Evaluated |
| Rank03 | ModelOP | Score86.8% | Percentile94.7% | Participants39 | EvidenceC | Evaluated |
| Rank04 | ModelOP | Score84.3% | Percentile92.1% | Participants39 | EvidenceC | Evaluated |
| Rank05 | ModelST | Score84.0% | Percentile89.5% | Participants39 | EvidenceC | Evaluated |
| Rank06 | ModelCO | Score80.6% | Percentile86.8% | Participants39 | EvidenceC | Evaluated |
| Rank07 | ModelMA | Score74.9% | Percentile84.2% | Participants39 | EvidenceC | Evaluated |
| Rank08 | ModelME | Score73.7% | Percentile81.6% | Participants39 | EvidenceC | Evaluated |
| Rank09 | ModelOP | Score73.1% | Percentile79.0% | Participants39 | EvidenceC | Evaluated |
| Rank10 | ModelOP | Score72.3% | Percentile76.3% | Participants39 | EvidenceC | Evaluated |
| Rank11 | ModelOP | Score72.2% | Percentile73.7% | Participants39 | EvidenceC | Evaluated |
| Rank12 | ModelOP | Score71.8% | Percentile71.0% | Participants39 | EvidenceC | Evaluated |
| Rank13 | ModelAC | Score71.4% | Percentile68.4% | Participants39 | EvidenceC | Evaluated |
| Rank14 | ModelME | Score70.7% | Percentile65.8% | Participants39 | EvidenceC | Evaluated |
| Rank15 | ModelMA | Score69.4% | Percentile63.2% | Participants39 | EvidenceC | Evaluated |
| Rank16 | ModelXA | Score69.0% | Percentile60.5% | Participants39 | EvidenceC | Evaluated |
| Rank17 | ModelGO | Score68.1% | Percentile57.9% | Participants39 | EvidenceC | Evaluated |
| Rank18 | ModelXA | Score68.1% | Percentile55.3% | Participants39 | EvidenceC | Evaluated |
| Rank19 | ModelAC | Score67.9% | Percentile52.6% | Participants39 | EvidenceC | Evaluated |
| Rank20 | ModelAN | Score67.7% | Percentile50.0% | Participants39 | EvidenceC | Evaluated |
| Rank21 | ModelMA | Score67.1% | Percentile47.4% | Participants39 | EvidenceC | Evaluated |
| Rank22 | ModelGO | Score65.8% | Percentile44.7% | Participants39 | EvidenceC | Evaluated |
| Rank23 | ModelOP | Score63.8% | Percentile42.1% | Participants39 | EvidenceC | Evaluated |
| Rank24 | ModelDE | Score62.8% | Percentile39.5% | Participants39 | EvidenceC | Evaluated |
| Rank25 | ModelMI | Score62.4% | Percentile36.8% | Participants39 | EvidenceC | Evaluated |
| Rank26 | ModelOP | Score61.4% | Percentile34.2% | Participants39 | EvidenceC | Evaluated |
| Rank27 | ModelDE | Score60.7% | Percentile31.6% | Participants39 | EvidenceC | Evaluated |
| Rank28 | ModelMA | Score58.0% | Percentile28.9% | Participants39 | EvidenceC | Evaluated |
| Rank29 | ModelME | Score57.3% | Percentile26.3% | Participants39 | EvidenceC | Evaluated |
| Rank30 | ModelOP | Score56.7% | Percentile23.7% | Participants39 | EvidenceC | Evaluated |
| Rank31 | ModelOP | Score56.2% | Percentile21.1% | Participants39 | EvidenceC | Evaluated |
| Rank32 | ModelGO | Score54.7% | Percentile18.4% | Participants39 | EvidenceC | Evaluated |
| Rank33 | ModelDE | Score53.6% | Percentile15.8% | Participants39 | EvidenceC | Evaluated |
| Rank34 | ModelXA | Score52.8% | Percentile13.2% | Participants39 | EvidenceC | Evaluated |
| Rank35 | ModelXA | Score52.8% | Percentile10.5% | Participants39 | EvidenceC | Evaluated |
| Rank36 | ModelME | Score51.5% | Percentile7.9% | Participants39 | EvidenceC | Evaluated |
| Rank37 | ModelGO | Score46.6% | Percentile5.3% | Participants39 | EvidenceB | Evaluated |
| Rank38 | ModelMI | Score43.9% | Percentile2.6% | Participants39 | EvidenceC | Evaluated |
| Rank39 | ModelOP | Score0.0% | Percentile0.0% | Participants39 | EvidenceB | Evaluated |
MathVista Highlights
The leading models and scores on this benchmark.
MathVista Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for MathVista
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
Ranking basisThis mathvista 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.
- 01BYSeed 2.1 ProByteDance
- Score
- 90.7%
Strengths
- Ranks #1 of 39 compared models
- 100th percentile on this benchmark
- C evidence result
Considerations
- This result measures MathVista, not total model capability
- 02BYSeed 2.1 TurboByteDance
- Score
- 90.5%
Strengths
- Ranks #2 of 39 compared models
- 97th percentile on this benchmark
- C evidence result
Considerations
- This result measures MathVista, not total model capability
- 03OPo3OpenAI
- Score
- 86.8%
- Price
- $2.0 input / $8.0 output per 1M tokens
- Speed
- Up to 50 tok/s via OpenAI
Strengths
- Ranks #3 of 39 compared models
- 95th percentile on this benchmark
- C evidence result
Considerations
- This result measures MathVista, not total model capability
- 04OPo4-miniOpenAI
- Score
- 84.3%
- Price
- $1.1 input / $4.4 output per 1M tokens
- Speed
- Up to 115 tok/s via OpenAI
Strengths
- Ranks #4 of 39 compared models
- 92th percentile on this benchmark
- C evidence result
Considerations
- This result measures MathVista, not total model capability
- 05STStep3-VL-10BStepFun
- Score
- 84.0%
Strengths
- Ranks #5 of 39 compared models
- 89th percentile on this benchmark
- C evidence result
Considerations
- This result measures MathVista, not total model capability
Selection summary
Best AI Models for MathVista
Seed 2.1 Pro currently leads MathVista with 90.7%. 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 MathVista?
What MathVista measures and how its scores work.
MathVista evaluates mathematical reasoning of foundation models in visual contexts. It consists of 6,141 examples derived from 28 existing multimodal datasets and 3 newly created datasets (IQTest, FunctionQA, and PaperQA), combining challenges from diverse mathematical and visual tasks to assess models' ability to understand complex figures and perform rigorous reasoning.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- MathVista
- Modality
- multimodal
- Primary category
- math
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- mathvista|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about MathVista.
Which model scores highest on MathVista?
Seed 2.1 Pro is currently ranked first with 90.7%.
What does MathVista measure?
MathVista evaluates mathematical reasoning of foundation models in visual contexts. It consists of 6,141 examples derived from 28 existing multimodal datasets and 3 newly created datasets (IQTest, FunctionQA, and PaperQA), combining challenges from diverse mathematical and visual tasks to assess models' ability to understand complex figures and perform rigorous reasoning.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
39 model results are currently shown.
Does this benchmark affect the overall score?
Yes. This benchmark can contribute to the current LLMBoard capability score.