Open LMM Reasoning - WeMath - Position: leaderboard
The "WeMath - Position" 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 | InternVL2.5-78B | 81.4 | #431 |
| 2 | Doubao-1.5-Pro | 81.4 | #287 |
| 3 | Gemini 1.5 Pro (002) | 78.1 | #349 |
| 4 | Qwen 2 VL 72B | 76.1 | #460 |
| 5 | InternVL3-78B | 74.7 | #352 |
| 6 | InternVL3-38B | 73.6 | #405 |
| 7 | GPT-4o ChatGPT | 73 | |
| 8 | Taichu-VLR-7B | 72.4 | #526 |
| 9 | Gemini 2.0 Flash | 71.7 | #342 |
| 10 | InternVL3-14B | 71.7 | #505 |
| 11 | Claude 3.7 Sonnet | 69.1 | #249 |
| 12 | GPT-4.1 Nano | 68.8 | #731 |
| 13 | GPT-4.1 (2025-04-14) | 68.2 | #250 |
| 14 | Gemma 3 27B | 68.1 | #607 |
| 15 | Qwen 2.5 VL 7B | 68 | #643 |
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-position · How It Works · Data refreshed daily, snapshot 2026-10-20.