BOTTLED - RAID: leaderboard
Metric: AUROC (0-1). Source: arxiv.org. Saturation forecast: Not forecast. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 0.82 |
| 2 | Qwen 3.8 Max | 0.77 |
| 3 | GLM-5.3 Flash | 0.74 |
| 4 | GPT-5.6 Terra | 0.73 |
| 5 | Claude Opus 5 | 0.71 |
| 6 | Qwen 3.8 Flash | 0.71 |
| 7 | Claude Sonnet 5 | 0.58 |
| 8 | Gemini 3.1 Flash Lite | 0.54 |
| 9 | GLM-5.3 | 0.52 |
| 10 | GPT-5.6 Sol | 0.5 |
Interactive version: theaggregate.ai/benchmark?slug=bottled-raid · How It Works · Data refreshed daily, snapshot 2026-10-09.