Benchmark
MathVista-Mini
MathVista-Mini is a smaller version of the MathVista benchmark that evaluates mathematical reasoning in visual contexts. It consists of examples derived…
- Modality
- multimodal
- Categories
- math, multimodal, vision
- Openness
- unknown
- Source
- llm_stats
- Reported scores
- 24
Reported scores
llm_stats
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| Gemma 3 12B | 0.629 | 2025-03-12 | evidence | |
| Gemma 3 27B | 0.676 | 2025-03-12 | evidence | |
| Gemma 3 4B | 0.5 | 2025-03-12 | evidence | |
| Kimi K2.5 | Moonshot AI | 0.901 | 2026-01-27 | evidence |
| LFM2.5-VL-3B | Liquid AI | 0.685 | 2026-08-12 | evidence |
| Qwen2-VL-72B-Instruct | Qwen | 0.705 | 2024-08-29 | evidence |
| Qwen2.5 VL 32B Instruct | Qwen | 0.747 | 2025-02-28 | evidence |
| Qwen2.5 VL 72B Instruct | Qwen | 0.748 | 2025-01-26 | evidence |
| Qwen2.5 VL 7B Instruct | Qwen | 0.682 | 2025-01-26 | evidence |
| Qwen3 VL 235B A22B Instruct | Qwen | 0.849 | 2025-09-22 | evidence |
| Qwen3 VL 235B A22B Thinking | Qwen | 0.858 | 2025-09-22 | evidence |
| Qwen3 VL 30B A3B Instruct | Qwen | 0.801 | 2025-09-22 | evidence |
| Qwen3 VL 30B A3B Thinking | Qwen | 0.819 | 2025-09-22 | evidence |
| Qwen3 VL 32B Instruct | Qwen | 0.838 | 2025-09-22 | evidence |
| Qwen3 VL 32B Thinking | Qwen | 0.859 | 2025-09-22 | evidence |
| Qwen3 VL 4B Instruct | Qwen | 0.737 | 2025-09-22 | evidence |
| Qwen3 VL 4B Thinking | Qwen | 0.795 | 2025-09-22 | evidence |
| Qwen3 VL 8B Instruct | Qwen | 0.772 | 2025-09-22 | evidence |
| Qwen3 VL 8B Thinking | Qwen | 0.814 | 2025-09-22 | evidence |
| Qwen3.5-122B-A10B | Qwen | 0.874 | 2026-02-24 | evidence |
| Qwen3.5-27B | Qwen | 0.878 | 2026-02-24 | evidence |
| Qwen3.5-35B-A3B | Qwen | 0.862 | 2026-02-24 | evidence |
| Qwen3.6-27B | Qwen | 0.874 | 2026-04-21 | evidence |
| Qwen3.6-35B-A3B | Qwen | 0.864 | 2026-04-16 | evidence |
Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.