Open LMM Reasoning - WeMath - Two-step(S2) — leaderboard

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

Top models

#ModelScore
1Doubao-1.5-Pro78.6
2GPT-4o ChatGPT69.2
3GPT-4.1 (2025-04-14)68.9
4Claude 3.7 Sonnet68.6
5InternVL3-38B67.2
6Gemini 1.5 Pro (002)65.8
7InternVL3-78B62.8
8InternVL3-14B62.8
9Gemini 2.0 Flash61.1
10Claude 3.5 Sonnet (20241022)60.6
11QVQ-72B-Preview59.2
12InternVL2.5-78B58.1
13InternVL3-8B55.8
14Gemma 3 27B55.8
15Grok 2 (1212)55.3

Interactive version: theaggregate.ai/benchmark?slug=open-lmm-reasoning-wemath-two-step-s2 · How the rankings work · Data refreshed daily, snapshot 2026-07-22.