Open LMM Reasoning - WeMath - RoteMemorization (Loose): leaderboard
The "WeMath - RoteMemorization (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. 83 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4.1 Mini | 91.7 | #355 |
| 2 | Llama 3.2 11B Instruct | 36 | #1126 |
| 3 | Qwen 2 VL 2B | 22.3 | #1220 |
| 4 | InternVL2.5-2B | 19.3 | #1249 |
| 5 | Aquila-VL-2B | 13.5 | #1238 |
| 6 | Qwen 2 VL 7B | 12.5 | #750 |
| 7 | Gemma 3 4B | 10.3 | #1089 |
| 8 | GPT-4.1 Nano | 7.8 | #731 |
| 9 | InternVL2.5-8B-BoN-8 | 7.8 | #639 |
| 10 | URSA-8B | 7.5 | #811 |
| 11 | VLM-R1-3B-Math-0305 | 7.5 | #743 |
| 12 | VLAA-Thinker-Qwen2.5VL-3B | 6.9 | #720 |
| 13 | Grok 2 (1212) | 6.5 | #496 |
| 14 | Moonshot v1 8K | 6.3 | #617 |
| 15 | Gemma 3 12B | 6.2 | #671 |
Interactive version: theaggregate.ai/benchmark?slug=open-lmm-reasoning-wemath-rotememorization-loose · How It Works · Data refreshed daily, snapshot 2026-10-20.