Open LMM Reasoning - WeMath — leaderboard

Metric: Accuracy (%). Source: huggingface.co. 83 models tracked.

Top models

#ModelScore
1Doubao-1.5-Pro65.7
2GPT-4.1 (2025-04-14)55.5
3GPT-4o ChatGPT50.6
4Gemini 1.5 Pro (002)50.5
5InternVL3-38B49.3
6Claude 3.7 Sonnet49.3
7Gemini 2.0 Flash47.4
8InternVL3-78B46.1
9Claude 3.5 Sonnet (20241022)44
10InternVL3-14B41.3
11Gemma 3 27B40.7
12InternVL2.5-78B39.8
13InternVL3-8B39.5
14QVQ-72B-Preview39
15Grok 2 (1212)38.1

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