Open LMM Reasoning - DynaMath — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 100 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4o ChatGPT | 48.5 |
| 2 | GPT-4.1 Mini | 47.7 |
| 3 | Doubao-1.5-Pro | 44.9 |
| 4 | GPT-4.1 (2025-04-14) | 43.3 |
| 5 | Gemini 2.0 Flash | 42.1 |
| 6 | Claude 3.7 Sonnet | 39.7 |
| 7 | Claude 3.5 Sonnet (20241022) | 35.7 |
| 8 | InternVL3-78B | 35.1 |
| 9 | InternVL3-38B | 34.7 |
| 10 | Gemini 1.5 Pro (002) | 31.5 |
| 11 | QVQ-72B-Preview | 30.7 |
| 12 | InternVL3-14B | 30.3 |
| 13 | Gemma 3 27B | 27.5 |
| 14 | InternVL3-8B | 25.7 |
| 15 | Grok 2 (1212) | 25 |
Interactive version: theaggregate.ai/benchmark?slug=open-lmm-reasoning-dynamath · How the rankings work · Data refreshed daily, snapshot 2026-07-22.