Benchmark
MathVista
MathVista evaluates mathematical reasoning of foundation models in visual contexts. It consists of 6,141 examples derived from 28 existing multimodal…
- Released
- 2023-10-03
- Modality
- multimodal
- Categories
- math, multimodal, vision
- Openness
- open
- Source
- llm_stats
- Reported scores
- 39
Reported scores
llm_stats
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| Claude 3.5 Sonnet | Anthropic | 0.677 | 2024-10-22 | evidence |
| Command A+ | Cohere | 0.806 | 2026-05-20 | evidence |
| DeepSeek VL2 | DeepSeek | 0.628 | 2024-12-13 | evidence |
| DeepSeek VL2 Small | DeepSeek | 0.607 | 2024-12-13 | evidence |
| DeepSeek VL2 Tiny | DeepSeek | 0.536 | 2024-12-13 | evidence |
| Gemini 1.0 Pro | 0.466 | 2024-02-15 | evidence | |
| Gemini 1.5 Flash | 0.658 | 2024-05-01 | evidence | |
| Gemini 1.5 Flash 8B | 0.547 | 2024-03-15 | evidence | |
| Gemini 1.5 Pro | 0.681 | 2024-05-01 | evidence | |
| GPT-3.5 Turbo | OpenAI | 0.0 | 2023-03-21 | evidence |
| GPT-4.1 | OpenAI | 0.722 | 2025-04-14 | evidence |
| GPT-4.1 mini | OpenAI | 0.731 | 2025-04-14 | evidence |
| GPT-4.1 nano | OpenAI | 0.562 | 2025-04-14 | evidence |
| GPT-4.5 | OpenAI | 0.723 | 2025-02-27 | evidence |
| GPT-4o | OpenAI | 0.638 | 2024-05-13 | evidence |
| GPT-4o | OpenAI | 0.614 | 2024-08-06 | evidence |
| GPT-4o mini | OpenAI | 0.567 | 2024-07-18 | evidence |
| Grok-1.5 | xAI | 0.528 | 2024-03-28 | evidence |
| Grok-1.5V | xAI | 0.528 | 2024-04-12 | evidence |
| Grok-2 | xAI | 0.69 | 2024-08-13 | evidence |
| Grok-2 mini | xAI | 0.681 | 2024-08-13 | evidence |
| Kimi-k1.5 | Moonshot AI | 0.749 | 2025-01-20 | evidence |
| Llama 3.2 11B Instruct | Meta | 0.515 | 2024-09-25 | evidence |
| Llama 3.2 90B Instruct | Meta | 0.573 | 2024-09-25 | evidence |
| Llama 4 Maverick | Meta | 0.737 | 2025-04-05 | evidence |
| Llama 4 Scout | Meta | 0.707 | 2025-04-05 | evidence |
| Mistral Small 3.2 24B Instruct | Mistral | 0.6709 | 2025-06-20 | evidence |
| o1 | OpenAI | 0.718 | 2024-12-17 | evidence |
| o3 | OpenAI | 0.868 | 2025-04-16 | evidence |
| o4-mini | OpenAI | 0.843 | 2025-04-16 | evidence |
| Phi-3.5-vision-instruct | Microsoft | 0.439 | 2024-08-23 | evidence |
| Phi-4-multimodal-instruct | Microsoft | 0.624 | 2025-02-01 | evidence |
| Pixtral Large | Mistral | 0.694 | 2024-11-18 | evidence |
| Pixtral-12B | Mistral | 0.58 | 2024-09-17 | evidence |
| QvQ-72B-Preview | Qwen | 0.714 | 2024-12-25 | evidence |
| Qwen2.5-Omni-7B | Qwen | 0.679 | 2025-03-27 | evidence |
| Seed 2.1 Pro | ByteDance | 0.907 | 2026-06-24 | evidence |
| Seed 2.1 Turbo | ByteDance | 0.905 | 2026-06-24 | evidence |
| Step3-VL-10B | StepFun | 0.84 | 2026-01-15 | 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.