Open LMM Reasoning - DynaMath - Level-elementary school: leaderboard
Metric: Accuracy (%). Source: huggingface.co. 100 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4.1 Mini | 65.1 |
| 2 | GPT-4o ChatGPT | 58.7 |
| 3 | Doubao-1.5-Pro | 58.7 |
| 4 | GPT-4.1 (2025-04-14) | 55.6 |
| 5 | InternVL3-78B | 54 |
| 6 | Claude 3.5 Sonnet (20241022) | 52.4 |
| 7 | Claude 3.7 Sonnet | 52.4 |
| 8 | Gemini 2.0 Flash | 52.4 |
| 9 | InternVL3-38B | 47.6 |
| 10 | GPT-4.1 Nano | 44.4 |
| 11 | Gemini 1.5 Pro (002) | 41.3 |
| 12 | Taichu-VLR-7B | 41.3 |
| 13 | InternVL2.5-78B-CoT | 41.3 |
| 14 | InternVL3-8B | 39.7 |
| 15 | Grok 2 (1212) | 39.7 |
Interactive version: theaggregate.ai/benchmark?slug=open-lmm-reasoning-dynamath-level-elementary-school · How It Works · Data refreshed daily, snapshot 2026-09-05.