Open LMM Reasoning - WeMath - InadequateGeneralization (Loose): leaderboard
The "WeMath - InadequateGeneralization (Loose)" 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 | InternVL3-8B | 16.2 | #620 |
| 2 | Gemma 3 4B | 15.6 | #1089 |
| 3 | Taichu-VLR-3B | 15.4 | #744 |
| 4 | Claude 3.5 Sonnet (20241022) | 15.2 | #291 |
| 5 | Gemma 3 27B | 15.1 | #607 |
| 6 | Gemma 3 12B | 14.7 | #671 |
| 7 | Qwen 2.5 VL 7B | 14.5 | #643 |
| 8 | VLAA-Thinker-Qwen2.5VL-7B | 14.5 | #555 |
| 9 | Qwen 2 VL 72B | 13.7 | #460 |
| 10 | Moonshot v1 8K | 13.5 | #617 |
| 11 | VLAA-Thinker-Qwen2.5VL-3B | 13.5 | #720 |
| 12 | Qwen 2 VL 7B | 13.3 | #750 |
| 13 | VLM-R1-3B-Math-0305 | 13.1 | #743 |
| 14 | Gemini 2.0 Flash | 12.9 | #342 |
| 15 | Gemini 1.5 Pro (002) | 12.9 | #349 |
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-inadequategeneralization-loose · How It Works · Data refreshed daily, snapshot 2026-10-20.