Open LMM Reasoning - MathVista - STA — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 100 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | InternVL3-78B | 87 |
| 2 | QVQ-72B-Preview | 86.7 |
| 3 | GPT-4.1 Mini | 86.4 |
| 4 | InternVL3-38B | 85 |
| 5 | InternVL3-14B | 84.7 |
| 6 | Doubao-1.5-Pro | 84.4 |
| 7 | Qwen 2 VL 72B | 83.4 |
| 8 | GPT-4o ChatGPT | 83.1 |
| 9 | GPT-4.1 (2025-04-14) | 81.7 |
| 10 | InternVL2.5-78B | 81.4 |
| 11 | Ovis2-8B | 80.7 |
| 12 | Gemini 2.0 Flash | 80.1 |
| 13 | Claude 3.5 Sonnet (20241022) | 79.1 |
| 14 | InternVL3-8B | 78.4 |
| 15 | Gemini 1.5 Pro (002) | 77.7 |
Interactive version: theaggregate.ai/benchmark?slug=open-lmm-reasoning-mathvista-sta · How the rankings work · Data refreshed daily, snapshot 2026-07-22.