Open LMM Reasoning - MathVista - NUM: leaderboard
The "MathVista - NUM" 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 | Step-1o | 63.2 | |
| 2 | ChatGPT-4o-latest | 59.7 | |
| 3 | SenseNova | 56.2 | |
| 4 | Doubao-1.5-Pro | 56.2 | #287 |
| 5 | SAIL-VL-1.6-8B | 56.2 | |
| 6 | InternVL3-78B | 55.6 | #352 |
| 7 | Ovis2-34B | 54.2 | |
| 8 | Step-1.5V | 53.5 | |
| 9 | HunYuan-Standard-Vision | 53.5 | |
| 10 | InternVL2.5-78B-MPO | 52.8 |
Interactive version: theaggregate.ai/benchmark?slug=open-lmm-reasoning-mathvista-num · How It Works · Data refreshed daily, snapshot 2026-10-20.