DoGBench: leaderboard
Metric: Composite score (0-100), held-out split (117 items). Source: dogbench.ai. Saturation forecast: Around 2028. 7 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3.8 Max | 47.26 |
| 2 | GPT-5.6 Sol | 46.22 |
| 3 | GLM-5.2 | 44.16 |
| 4 | Kimi K2.7 Code | 43.88 |
| 5 | Claude Opus 4.8 | 41.16 |
| 6 | Claude Sonnet 4.6 | 40.02 |
| 7 | GPT-5.5 | 34.02 |
Interactive version: theaggregate.ai/benchmark?slug=dogbench · How It Works · Data refreshed daily, snapshot 2026-10-03.