Guesswork 2026-07: leaderboard

Metric: MAE (z-score units). Source: theaggregate.ai. 37 models tracked.

Top models

#ModelScore
1GLM-5.20.67
2Claude Opus 50.68
3Grok 4.60.73
4Gemini 3.1 Pro (Preview)0.77
5Claude Sonnet 50.81
6Gemini 2.5 Pro0.82
7DeepSeek V4 Pro0.82
8Qwen 3.7 Max0.87
9Grok 4.30.91
10Mistral Small 40.93
11Kimi K2.60.98
12GPT-5.50.98
13codestral-25081
14Claude Opus 4.81
15Mistral Medium 3.11.01

Interactive version: theaggregate.ai/benchmark?slug=guesswork-2026-07 · How It Works · Data refreshed daily, snapshot 2026-09-05.