Open LMM Reasoning - MathVision - transformation geometry: leaderboard
Metric: Accuracy (%). Source: huggingface.co. 100 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Doubao-1.5-Pro | 42.3 |
| 2 | Claude 3.5 Sonnet (20241022) | 35.7 |
| 3 | GPT-4.1 Mini | 35.1 |
| 4 | Gemini 2.0 Flash | 33.9 |
| 5 | GPT-4.1 (2025-04-14) | 33.9 |
| 6 | Claude 3.7 Sonnet | 33.3 |
| 7 | Gemini 1.5 Pro (002) | 31 |
| 8 | Gemma 3 27B | 31 |
| 9 | InternVL3-78B | 30.4 |
| 10 | Qwen 2 VL 72B | 30.4 |
| 11 | GPT-4o ChatGPT | 29.8 |
| 12 | InternVL3-38B | 29.2 |
| 13 | VLAA-Thinker-Qwen2.5VL-7B | 28 |
| 14 | URSA-8B | 27.4 |
| 15 | Taichu-VLR-3B | 27.4 |
Interactive version: theaggregate.ai/benchmark?slug=open-lmm-reasoning-mathvision-transformation-geometry · How It Works · Data refreshed daily, snapshot 2026-09-05.