Open LMM Reasoning - MathVerse - Property — leaderboard

Metric: Accuracy (%). Source: huggingface.co. 100 models tracked.

Top models

#ModelScore
1Gemini 1.5 Pro (002)71.8
2GPT-4.1 (2025-04-14)62
3Doubao-1.5-Pro62
4Gemini 2.0 Flash59.2
5GPT-4o ChatGPT59.2
6GPT-4.1 Mini57.7
7InternVL3-78B56.3
8InternVL3-38B54.9
9Qwen 2 VL 72B53.5
10Claude 3.7 Sonnet53.5
11Claude 3.5 Sonnet (20241022)52.1
12InternVL2.5-78B52.1
13QVQ-72B-Preview50.7
14InternVL3-14B43.7
15Ovis2-8B39.4

Interactive version: theaggregate.ai/benchmark?slug=open-lmm-reasoning-mathverse-property · How the rankings work · Data refreshed daily, snapshot 2026-07-22.