Open LMM Reasoning - DynaMath - Subject-arithmetic; Avg: leaderboard
Metric: Accuracy (%). Source: huggingface.co. 100 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4.1 Mini | 73.8 |
| 2 | GPT-4o ChatGPT | 72.7 |
| 3 | GPT-4.1 (2025-04-14) | 71.2 |
| 4 | Doubao-1.5-Pro | 61.2 |
| 5 | InternVL3-38B | 54.2 |
| 6 | Claude 3.7 Sonnet | 53.5 |
| 7 | Claude 3.5 Sonnet (20241022) | 52.3 |
| 8 | Gemini 2.0 Flash | 51.2 |
| 9 | Gemini 1.5 Pro (002) | 49.6 |
| 10 | GPT-4.1 Nano | 49.6 |
| 11 | QVQ-72B-Preview | 48.8 |
| 12 | Taichu-VLR-7B | 48.8 |
| 13 | InternVL3-78B | 47.3 |
| 14 | InternVL2.5-78B-CoT | 46.9 |
| 15 | Gemma 3 27B | 46.5 |
Interactive version: theaggregate.ai/benchmark?slug=open-lmm-reasoning-dynamath-subject-arithmetic-avg · How It Works · Data refreshed daily, snapshot 2026-09-05.