Open LMM Reasoning - MathVista - GPS — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 100 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Doubao-1.5-Pro | 91.3 |
| 2 | InternVL3-78B | 86.1 |
| 3 | Gemini 2.0 Flash | 84.6 |
| 4 | Gemini 1.5 Pro (002) | 81.7 |
| 5 | Gemma 3 27B | 80.8 |
| 6 | InternVL3-38B | 80.3 |
| 7 | InternVL3-14B | 79.8 |
| 8 | Ovis2-8B | 78.4 |
| 9 | InternVL3-8B | 77.4 |
| 10 | QVQ-72B-Preview | 76.4 |
| 11 | Grok 2 (1212) | 74.5 |
| 12 | GPT-4o ChatGPT | 73.1 |
| 13 | InternVL2.5-78B | 72.1 |
| 14 | Claude 3.7 Sonnet | 72.1 |
| 15 | Gemma 3 12B | 71.6 |
Interactive version: theaggregate.ai/benchmark?slug=open-lmm-reasoning-mathvista-gps · How the rankings work · Data refreshed daily, snapshot 2026-07-22.