← Benchmarks
MathVision
MATH-Vision is a dataset designed to measure multimodal mathematical reasoning capabilities. It focuses on evaluating how well models can solve mathematical problems that require both visual understanding and mathematical reasoning, bridging the gap between visual and mathematical domains.
id mathvision · max 1 · 37 models reported
| # | Model | Score |
|---|---|---|
| 1 | Kimi K3 Moonshot AI · open | 0.98 |
| 2 | Qwen3.8 Flash Alibaba Cloud / Qwen Team | 0.96 |
| 3 | Qwen3.8-Flash-Next Alibaba Cloud / Qwen Team · open | 0.96 |
| 4 | Qwen3.8-27B Alibaba Cloud / Qwen Team · open | 0.95 |
| 5 | Seed 2.1 Pro ByteDance | 0.94 |
| 6 | Kimi K2.6 Moonshot AI · open | 0.93 |
| 7 | Seed 2.1 Turbo ByteDance | 0.93 |
| 8 | Qwen3.7-Plus Alibaba Cloud / Qwen Team | 0.90 |
| 9 | Qwen3.6 Plus Alibaba Cloud / Qwen Team | 0.88 |
| 10 | Qwen3.5-122B-A10B Alibaba Cloud / Qwen Team · open | 0.86 |
| 11 | Qwen3.5-27B Alibaba Cloud / Qwen Team · open | 0.86 |
| 12 | Gemma 4 31B Google · open | 0.86 |