Open LMM Reasoning - DynaMath - Level-elementary school; Avg: leaderboard
Metric: Accuracy (%). Source: huggingface.co. 100 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4.1 Mini | 78.1 |
| 2 | GPT-4o ChatGPT | 76.5 |
| 3 | GPT-4.1 (2025-04-14) | 74.9 |
| 4 | Doubao-1.5-Pro | 72.9 |
| 5 | Claude 3.7 Sonnet | 67.8 |
| 6 | InternVL3-38B | 67 |
| 7 | Gemini 2.0 Flash | 67 |
| 8 | InternVL3-78B | 66.7 |
| 9 | Claude 3.5 Sonnet (20241022) | 65.1 |
| 10 | Gemini 1.5 Pro (002) | 64.8 |
| 11 | InternVL2.5-78B-CoT | 63 |
| 12 | QVQ-72B-Preview | 61.4 |
| 13 | Taichu-VLR-7B | 61.4 |
| 14 | GPT-4.1 Nano | 60.2 |
| 15 | VLAA-Thinker-Qwen2.5VL-7B | 59.4 |
Interactive version: theaggregate.ai/benchmark?slug=open-lmm-reasoning-dynamath-level-elementary-school-avg · How It Works · Data refreshed daily, snapshot 2026-09-05.