Guesswork 2026-07: leaderboard
Guesswork standings for 2026-07: each contestant predicts newly scraped (model, benchmark) scores from day-before information only, scored as mean absolute error in z-score units — lower is better. The Aggregate's own row is excluded so the site's predictor cannot feed itself a score.
Metric: MAE (z-score units). Source: theaggregate.ai. Saturation forecast: Around October 2027. 37 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GLM-5.2 | 0.67 | #73 |
| 2 | Claude Opus 5 | 0.68 | #12 |
| 3 | Grok 4.6 | 0.73 | #26 |
| 4 | Gemini 3.1 Pro (Preview) | 0.77 | #58 |
| 5 | Claude Sonnet 5 | 0.81 | #57 |
| 6 | Gemini 2.5 Pro | 0.82 | #152 |
| 7 | DeepSeek V4 Pro | 0.82 | #102 |
| 8 | Qwen 3.7 Max | 0.87 | #75 |
| 9 | Grok 4.3 | 0.91 | #188 |
| 10 | Mistral Small 4 | 0.93 | #517 |
| 11 | Kimi K2.6 | 0.98 | #104 |
| 12 | GPT-5.5 | 0.98 | #31 |
| 13 | codestral-2508 | 1 | #531 |
| 14 | Claude Opus 4.8 | 1 | #37 |
| 15 | Mistral Medium 3.1 | 1.01 | #408 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=guesswork-2026-07 · How It Works · Data refreshed daily, snapshot 2026-10-20.