Open LMM Reasoning - DynaMath - Subject-solid geometry: leaderboard
The "DynaMath - Subject-solid geometry" 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 | 40 | #355 |
| 2 | Gemini 2.0 Flash | 40 | #342 |
| 3 | GPT-4.1 (2025-04-14) | 40 | #250 |
| 4 | Doubao-1.5-Pro | 40 | #287 |
| 5 | Claude 3.5 Sonnet (20241022) | 33.3 | #291 |
| 6 | Gemini 1.5 Pro (002) | 33.3 | #349 |
| 7 | Gemma 3 27B | 33.3 | #607 |
| 8 | Grok 2 (1212) | 33.3 | #496 |
| 9 | GPT-4o ChatGPT | 33.3 | |
| 10 | InternVL3-8B | 26.7 | #620 |
| 11 | Claude 3.7 Sonnet | 26.7 | #249 |
| 12 | Gemma 3 12B | 26.7 | #671 |
| 13 | Moonshot v1 8K | 26.7 | #617 |
| 14 | QVQ-72B-Preview | 26.7 | #440 |
| 15 | VLAA-Thinker-Qwen2.5VL-3B | 26.7 | #720 |
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-solid-geometry · How It Works · Data refreshed daily, snapshot 2026-10-20.