Open LMM Reasoning - MathVista - VQA: leaderboard
The "MathVista - VQA" 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 | 64.8 | |
| 2 | ChatGPT-4o-latest | 64.2 | |
| 3 | InternVL3-78B | 63.7 | #352 |
| 4 | Ovis2-34B | 62.6 | |
| 5 | Qwen2.5-VL-72B | 62.6 | |
| 6 | InternVL3-38B | 62.6 | #405 |
| 7 | Ovis2-4B | 62 | |
| 8 | Doubao-1.5-Pro | 61.5 | #287 |
| 9 | InternVL2.5-38B | 60.9 | |
| 10 | Ovis1.6-Gemma2-27B | 60.3 |
Interactive version: theaggregate.ai/benchmark?slug=open-lmm-reasoning-mathvista-vqa · How It Works · Data refreshed daily, snapshot 2026-10-20.