Open LMM Reasoning - MathVision - topology: leaderboard
Metric: Accuracy (%). Source: huggingface.co. 100 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4.1 Mini | 43.5 |
| 2 | InternVL2.5-78B | 39.1 |
| 3 | InternVL3-38B | 34.8 |
| 4 | InternVL2.5-2B | 34.8 |
| 5 | VLAA-Thinker-Qwen2.5VL-7B | 34.8 |
| 6 | InternVL3-8B | 30.4 |
| 7 | Gemma 3 12B | 30.4 |
| 8 | VLAA-Thinker-Qwen2.5VL-3B | 30.4 |
| 9 | InternVL3-78B | 26.1 |
| 10 | InternVL2-8B | 26.1 |
| 11 | Taichu-VLR-3B | 26.1 |
| 12 | Gemma 3 27B | 21.7 |
| 13 | Gemini 2.0 Flash | 21.7 |
| 14 | GPT-4.1 (2025-04-14) | 21.7 |
| 15 | Taichu-VLR-7B | 21.7 |
Interactive version: theaggregate.ai/benchmark?slug=open-lmm-reasoning-mathvision-topology · How It Works · Data refreshed daily, snapshot 2026-09-05.