Open LMM Reasoning - MathVerse - Angle: leaderboard
The "MathVerse - Angle" 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 | Doubao-1.5-Pro | 68.4 | #287 |
| 2 | Gemini 1.5 Pro (002) | 55.4 | #349 |
| 3 | VLAA-Thinker-Qwen2.5VL-7B | 54.9 | #555 |
| 4 | Gemini 2.0 Flash | 52.3 | #342 |
| 5 | QVQ-72B-Preview | 49.7 | #440 |
| 6 | InternVL3-78B | 49.2 | #352 |
| 7 | Taichu-VLR-7B | 49.2 | #526 |
| 8 | Claude 3.5 Sonnet (20241022) | 47.7 | #291 |
| 9 | InternVL2.5-78B | 46.6 | #431 |
| 10 | InternVL3-38B | 46.1 | #405 |
| 11 | GPT-4.1 (2025-04-14) | 45.1 | #250 |
| 12 | GPT-4o ChatGPT | 44.6 | |
| 13 | InternVL3-14B | 43 | #505 |
| 14 | Qwen 2 VL 72B | 41.5 | #460 |
| 15 | Qwen 2.5 VL 7B | 41.5 | #643 |
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-mathverse-angle · How It Works · Data refreshed daily, snapshot 2026-10-20.