Open LMM Reasoning - MathVision - topology: leaderboard
The "MathVision - topology" 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 | GPT-4.1 Mini | 43.5 | #355 |
| 2 | InternVL2.5-78B | 39.1 | #431 |
| 3 | InternVL3-38B | 34.8 | #405 |
| 4 | InternVL2.5-2B | 34.8 | #1249 |
| 5 | VLAA-Thinker-Qwen2.5VL-7B | 34.8 | #555 |
| 6 | InternVL3-8B | 30.4 | #620 |
| 7 | MiniCPM-V-2.6 | 30.4 | #836 |
| 8 | Gemma 3 12B | 30.4 | #671 |
| 9 | VLAA-Thinker-Qwen2.5VL-3B | 30.4 | #720 |
| 10 | InternVL3-78B | 26.1 | #352 |
| 11 | InternVL2-8B | 26.1 | #817 |
| 12 | Taichu-VLR-3B | 26.1 | #744 |
| 13 | Gemini 2.0 Flash | 21.7 | #342 |
| 14 | GPT-4.1 (2025-04-14) | 21.7 | #250 |
| 15 | InternVL3-14B | 21.7 | #505 |
Interactive version: theaggregate.ai/benchmark?slug=open-lmm-reasoning-mathvision-topology · How It Works · Data refreshed daily, snapshot 2026-10-20.