PASTABench: leaderboard
Metric: Perfect interruption rate (%; share of the 1,139 risky multi-turn agent trajectories on which the monitor first interrupts inside the optimal intervention window, at or after the earliest-signal turn and before the trigger turn, and names the correct level-2 risk sub-category; zero-shot, greedy decoding, stepwise auto-stop protocol). Source: arxiv.org. Saturation forecast: Around 2032. 16 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4o | 40.74 |
| 2 | Gemini 3 Pro | 39.42 |
| 3 | Llama 4 Maverick | 33.36 |
| 4 | Llama 4 Scout | 30.99 |
| 5 | DeepSeek R1 | 30.47 |
| 6 | Grok 3 | 29.94 |
| 7 | Llama 3.3 70B Instruct | 24.06 |
| 8 | Gemini 2.5 Pro | 23.79 |
| 9 | QwQ-32B | 22.65 |
Interactive version: theaggregate.ai/benchmark?slug=pastabench · How It Works · Data refreshed daily, snapshot 2026-09-26.