Guesswork — leaderboard
This site's own live benchmark for predicting benchmark scores: frontier and budget LLMs forecast newly published (model, benchmark) results a day before they land, scored in per-benchmark z-units (MAE, lower is better) against the same-information statistical baseline The Aggregate.
Metric: MAE (z-score units). Source: aibenchmarks.dev. Status: saturated. 21 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GLM-5.2 | 0.66 |
| 2 | Gemini 3.1 Pro (Preview) | 0.76 |
| 3 | Gemini 2.5 Pro | 0.8 |
| 4 | Claude Sonnet 5 | 0.8 |
| 5 | DeepSeek V4 Pro | 0.8 |
| 6 | Qwen 3.7 Max | 0.85 |
| 7 | Grok 4.3 | 0.89 |
| 8 | Kimi K2.6 | 0.96 |
| 9 | GPT-5.5 | 0.96 |
| 10 | Claude Opus 4.8 | 0.98 |
| 11 | GPT-5.6 Terra | 1.01 |
| 12 | Ling-2.6-flash | 1.01 |
| 13 | GPT-5.1 | 1.02 |
| 14 | Qwen 3.5 9B | 1.02 |
| 15 | GPT-OSS-20B | 1.03 |
Interactive version: theaggregate.ai/benchmark?slug=guesswork · How the rankings work · Data refreshed daily, snapshot 2026-07-22.