math benchmark
MathVista-Mini Leaderboard
MathVista-Mini is a smaller version of the MathVista benchmark that evaluates mathematical reasoning in visual contexts. It consists of examples derived from multimodal datasets involving mathematics, combining challenges from diverse mathematical and visual tasks to assess foundation models' ability to solve problems requiring both visual understanding and mathematical reasoning.
Updated Aug 17, 2026
Models24
Model coverage24
MetricScore
EvidenceB
MathVista-Mini Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelMA | Score90.1% | Percentile100.0% | Participants24 | EvidenceC | Evaluated |
| Rank02 | ModelAC | Score87.8% | Percentile95.7% | Participants24 | EvidenceC | Evaluated |
| Rank03 | ModelAC | Score87.4% | Percentile91.3% | Participants24 | EvidenceC | Evaluated |
| Rank04 | ModelAC | Score87.4% | Percentile87.0% | Participants24 | EvidenceC | Evaluated |
| Rank05 | ModelAC | Score86.4% | Percentile82.6% | Participants24 | EvidenceC | Evaluated |
| Rank06 | ModelAC | Score86.2% | Percentile78.3% | Participants24 | EvidenceC | Evaluated |
| Rank07 | ModelAC | Score85.9% | Percentile73.9% | Participants24 | EvidenceC | Evaluated |
| Rank08 | ModelAC | Score85.8% | Percentile69.6% | Participants24 | EvidenceC | Evaluated |
| Rank09 | ModelAC | Score84.9% | Percentile65.2% | Participants24 | EvidenceC | Evaluated |
| Rank10 | ModelAC | Score83.8% | Percentile60.9% | Participants24 | EvidenceC | Evaluated |
| Rank11 | ModelAC | Score81.9% | Percentile56.5% | Participants24 | EvidenceC | Evaluated |
| Rank12 | ModelAC | Score81.4% | Percentile52.2% | Participants24 | EvidenceC | Evaluated |
| Rank13 | ModelAC | Score80.1% | Percentile47.8% | Participants24 | EvidenceC | Evaluated |
| Rank14 | ModelAC | Score79.5% | Percentile43.5% | Participants24 | EvidenceC | Evaluated |
| Rank15 | ModelAC | Score77.2% | Percentile39.1% | Participants24 | EvidenceC | Evaluated |
| Rank16 | ModelAC | Score74.8% | Percentile34.8% | Participants24 | EvidenceC | Evaluated |
| Rank17 | ModelAC | Score74.7% | Percentile30.4% | Participants24 | EvidenceC | Evaluated |
| Rank18 | ModelAC | Score73.7% | Percentile26.1% | Participants24 | EvidenceC | Evaluated |
| Rank19 | ModelAC | Score70.5% | Percentile21.7% | Participants24 | EvidenceC | Evaluated |
| Rank20 | ModelLA | Score68.5% | Percentile17.4% | Participants24 | EvidenceC | Evaluated |
| Rank21 | ModelAC | Score68.2% | Percentile13.0% | Participants24 | EvidenceC | Evaluated |
| Rank22 | ModelGO | Score67.6% | Percentile8.7% | Participants24 | EvidenceC | Evaluated |
| Rank23 | ModelGO | Score62.9% | Percentile4.3% | Participants24 | EvidenceC | Evaluated |
| Rank24 | ModelGO | Score50.0% | Percentile0.0% | Participants24 | EvidenceC | Evaluated |
MathVista-Mini Highlights
The leading models and scores on this benchmark.
MathVista-Mini Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for MathVista-Mini
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
Ranking basisThis mathvista-mini 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.
- 01MAKimi K2.5Moonshot AI
- Score
- 90.1%
- Price
- $0.60 input / $3.0 output per 1M tokens
Strengths
- Ranks #1 of 24 compared models
- 100th percentile on this benchmark
- C evidence result
Considerations
- This result measures MathVista-Mini, not total model capability
- 02ACQwen3.5-27BAlibaba Cloud / Qwen Team
- Score
- 87.8%
- Price
- $0.30 input / $2.4 output per 1M tokens
- Speed
- Up to 6.7 tok/s via Novita
Strengths
- Ranks #2 of 24 compared models
- 96th percentile on this benchmark
- C evidence result
Considerations
- This result measures MathVista-Mini, not total model capability
- 03ACQwen3.5-122B-A10BAlibaba Cloud / Qwen Team
- Score
- 87.4%
- Price
- $0.40 input / $3.2 output per 1M tokens
Strengths
- Ranks #3 of 24 compared models
- 91th percentile on this benchmark
- C evidence result
Considerations
- This result measures MathVista-Mini, not total model capability
- 04ACQwen3.6-27BAlibaba Cloud / Qwen Team
- Score
- 87.4%
- Price
- $0.60 input / $3.6 output per 1M tokens
- Speed
- Up to 6.1 tok/s via Novita
Strengths
- Ranks #4 of 24 compared models
- 87th percentile on this benchmark
- C evidence result
Considerations
- This result measures MathVista-Mini, not total model capability
- 05ACQwen3.6-35B-A3BAlibaba Cloud / Qwen Team
- Score
- 86.4%
- Price
- $0.25 input / $1.5 output per 1M tokens
Strengths
- Ranks #5 of 24 compared models
- 83th percentile on this benchmark
- C evidence result
Considerations
- This result measures MathVista-Mini, not total model capability
Selection summary
Best AI Models for MathVista-Mini
Kimi K2.5 currently leads MathVista-Mini with 90.1%. 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-Mini?
What MathVista-Mini measures and how its scores work.
MathVista-Mini is a smaller version of the MathVista benchmark that evaluates mathematical reasoning in visual contexts. It consists of examples derived from multimodal datasets involving mathematics, combining challenges from diverse mathematical and visual tasks to assess foundation models' ability to solve problems requiring both visual understanding and mathematical reasoning.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- MathVista-Mini
- Modality
- multimodal
- Primary category
- math
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- mathvista-mini|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about MathVista-Mini.
Which model scores highest on MathVista-Mini?
Kimi K2.5 is currently ranked first with 90.1%.
What does MathVista-Mini measure?
MathVista-Mini is a smaller version of the MathVista benchmark that evaluates mathematical reasoning in visual contexts. It consists of examples derived from multimodal datasets involving mathematics, combining challenges from diverse mathematical and visual tasks to assess foundation models' ability to solve problems requiring both visual understanding and mathematical reasoning.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
24 model results are currently shown.
Does this benchmark affect the overall score?
Yes. This benchmark can contribute to the current LLMBoard capability score.