MathVista — leaderboard
MathVista evaluates mathematical reasoning over visual contexts, including figure QA, geometry, word problems, textbook QA, and visual QA.
Metric: Accuracy (%). Source: huggingface.co. Status: saturation imminent. 284 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5 (2025-08-07) | 81.9 |
| 2 | Gemini 2.5 Pro | 80.9 |
| 3 | GPT-5 Mini (2025-08-07) | 79.2 |
| 4 | InternVL3-78B | 79 |
| 5 | Doubao-1.5-Pro | 78.6 |
| 6 | InternVL3-38B | 76.3 |
| 7 | InternVL3-14B | 74.4 |
| 8 | GPT-5 Nano | 73.1 |
| 9 | Ovis2-8B | 71.8 |
| 10 | GPT-4o ChatGPT | 71.6 |
| 11 | GPT-4.1 Mini | 70.9 |
| 12 | InternVL2.5-78B | 70.6 |
| 13 | InternVL3-8B | 70.5 |
| 14 | GPT-4.5 | 70.5 |
| 15 | Gemini 2.0 Flash | 70.4 |
Interactive version: theaggregate.ai/benchmark?slug=mathvista · How the rankings work · Data refreshed daily, snapshot 2026-07-22.