MathConstraint — leaderboard

Metric: Accuracy (self-reported). Source: benchmarklist.com. 12 models tracked.

Top models

#ModelScore
1GPT-5.566.9
2Grok 4.2058.1
3Gemini 3.1 Pro (Preview)51.7
4Qwen 3.6 Plus47.4
5Claude Opus 4.742.6
6GPT-OSS-120B39.5
7DeepSeek V4 Pro39.5
8DeepSeek V4 Flash34
9Gemini 3.1 Flash Lite24.9
10Claude Sonnet 4.618.5
11Llama 3.3 70B Instruct8.5

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