Open LMM Reasoning - MathVista: leaderboard
The "MathVista" subtask of Open LMM Reasoning. A task from the Open LMM Reasoning leaderboard, which evaluates large multimodal models on multi-step visual, mathematical and logical reasoning problems.
Metric: Accuracy (%). Source: huggingface.co. 100 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | InternVL3-78B | 79 | #352 |
| 2 | Doubao-1.5-Pro | 78.6 | #287 |
| 3 | SenseNova | 78.4 | |
| 4 | InternVL2.5-78B-MPO | 76.6 | |
| 5 | InternVL3-38B | 76.3 | #405 |
| 6 | Ovis2-34B | 76.1 | |
| 7 | MUG-U-7B | 74.8 | |
| 8 | Step-1o | 74.7 | |
| 9 | InternVL3-14B | 74.4 | #505 |
| 10 | Qwen2.5-VL-72B | 74.2 |
Interactive version: theaggregate.ai/benchmark?slug=open-lmm-reasoning-mathvista · How It Works · Data refreshed daily, snapshot 2026-10-20.