Open LMM Reasoning - WeMath - Score (Loose) — leaderboard

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

Top models

#ModelScore
1Doubao-1.5-Pro80.2
2GPT-4.1 (2025-04-14)71
3GPT-4o ChatGPT69.6
4Claude 3.7 Sonnet67.8
5InternVL3-38B67.2
6Gemini 1.5 Pro (002)66.9
7Gemini 2.0 Flash66.1
8Claude 3.5 Sonnet (20241022)63.8
9InternVL3-78B63
10InternVL3-14B62.1
11QVQ-72B-Preview60.7
12InternVL2.5-78B59.2
13InternVL3-8B58.8
14Gemma 3 27B58.4
15Grok 2 (1212)55.6

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