Open LMM Reasoning - WeMath - Angles and Length: leaderboard
The "WeMath - Angles and Length" 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. Saturation forecast: Estimated already saturated. 83 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Doubao-1.5-Pro | 66.3 | #287 |
| 2 | Gemini 1.5 Pro (002) | 61.8 | #349 |
| 3 | Gemini 2.0 Flash | 56.3 | #342 |
| 4 | Taichu-VLR-7B | 55.8 | #526 |
| 5 | GPT-4.1 (2025-04-14) | 55.1 | #250 |
| 6 | Gemma 3 27B | 53.7 | #607 |
| 7 | Taichu-VLR-3B | 53.7 | #744 |
| 8 | Qwen 2.5 VL 7B | 51.8 | #643 |
| 9 | VLAA-Thinker-Qwen2.5VL-3B | 51 | #720 |
| 10 | VLM-R1-3B-Math-0305 | 48.4 | #743 |
| 11 | VLAA-Thinker-Qwen2.5VL-7B | 47.7 | #555 |
| 12 | Moonshot v1 8K | 45.8 | #617 |
| 13 | Qwen 2 VL 72B | 45.1 | #460 |
| 14 | InternVL3-38B | 42.5 | #405 |
| 15 | GPT-4o ChatGPT | 42.5 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=open-lmm-reasoning-wemath-angles-and-length · How It Works · Data refreshed daily, snapshot 2026-10-20.