Guesswork 2026-08: leaderboard
Guesswork standings for 2026-08: 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 December 2027. 33 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 5 | 0.89 | #12 |
| 2 | Gemini 3.1 Pro (Preview) | 0.96 | #58 |
| 3 | Gemini 3.7 Flash | 0.97 | #23 |
| 4 | Grok 4.6 | 0.98 | #26 |
| 5 | Claude Sonnet 5 | 0.98 | #57 |
| 6 | Claude Opus 4.8 | 1 | #37 |
| 7 | GPT-5.5 | 1 | #31 |
| 8 | Ministral 3 14B | 1.02 | #650 |
| 9 | Kimi K2.6 | 1.03 | #104 |
| 10 | Gemma 3 4B (IT) | 1.05 | #992 |
| 11 | GPT-5.6 Luna | 1.05 | #68 |
| 12 | GPT-5.6 Terra | 1.06 | #41 |
| 13 | Mistral Small 3 | 1.08 | #627 |
| 14 | GPT-5.1 | 1.1 | #135 |
| 15 | Qwen 3.7 Max | 1.1 | #75 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=guesswork-2026-08 · How It Works · Data refreshed daily, snapshot 2026-10-20.