Open LMM Reasoning - MathVista - GPS: leaderboard
The "MathVista - GPS" 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 | Doubao-1.5-Pro | 91.3 | #287 |
| 2 | InternVL2.5-78B-MPO | 87 | |
| 3 | InternVL3-78B | 86.1 | #352 |
| 4 | Gemini-2.0-Pro | 85.6 | |
| 5 | SenseNova | 85.1 | |
| 6 | GLM-4v-Plus-20250111 | 85.1 | |
| 7 | Gemini-2.0-Flash | 84.6 | |
| 8 | InternVL2.5-38B-MPO | 84.1 | |
| 9 | Ovis2-34B | 83.7 | |
| 10 | MiniCPM-o-2.6 | 83.2 |
Interactive version: theaggregate.ai/benchmark?slug=open-lmm-reasoning-mathvista-gps · How It Works · Data refreshed daily, snapshot 2026-10-20.