Open LMM Reasoning - DynaMath - Subject-arithmetic; Avg: leaderboard
The "DynaMath - Subject-arithmetic; Avg" 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. 100 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4.1 Mini | 73.8 | #355 |
| 2 | GPT-4o ChatGPT | 72.7 | |
| 3 | GPT-4.1 (2025-04-14) | 71.2 | #250 |
| 4 | Doubao-1.5-Pro | 61.2 | #287 |
| 5 | InternVL3-38B | 54.2 | #405 |
| 6 | Claude 3.7 Sonnet | 53.5 | #249 |
| 7 | Claude 3.5 Sonnet (20241022) | 52.3 | #291 |
| 8 | Gemini 2.0 Flash | 51.2 | #342 |
| 9 | GPT-4.1 Nano | 49.6 | #731 |
| 10 | Gemini 1.5 Pro (002) | 49.6 | #349 |
| 11 | QVQ-72B-Preview | 48.8 | #440 |
| 12 | Taichu-VLR-7B | 48.8 | #526 |
| 13 | InternVL3-14B | 47.7 | #505 |
| 14 | InternVL3-78B | 47.3 | #352 |
| 15 | InternVL2.5-78B-CoT | 46.9 | #469 |
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-dynamath-subject-arithmetic-avg · How It Works · Data refreshed daily, snapshot 2026-10-20.