← Benchmarks
MathVista-Mini
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.
id mathvista-mini · max 1 · 26 models reported
| # | Model | Score |
|---|---|---|
| 1 | Kimi K2.5 Moonshot AI · open | 0.90 |
| 2 | Qwen3.5-27B Alibaba Cloud / Qwen Team · open | 0.88 |
| 3 | Qwen3.5-122B-A10B Alibaba Cloud / Qwen Team · open | 0.87 |
| 4 | Qwen3.6-27B Alibaba Cloud / Qwen Team · open | 0.87 |
| 5 | Qwen3.6-35B-A3B Alibaba Cloud / Qwen Team · open | 0.86 |
| 6 | Qwen3.5-35B-A3B Alibaba Cloud / Qwen Team · open | 0.86 |
| 7 | Qwen3 VL 32B Thinking Alibaba Cloud / Qwen Team · open | 0.86 |
| 8 | Qwen3 VL 235B A22B Thinking Alibaba Cloud / Qwen Team · open | 0.86 |
| 9 | Seed 2.0 Mini ByteDance | 0.85 |
| 10 | EXAONE 4.5 33B LG AI Research | 0.85 |