PreAct-Bench - Daily Life: leaderboard
Metric: Prefix Foresight F1 with the full prefix (PFF(100), binary F1 of the unethical class, 0-1 scaled to %): given an agent trajectory truncated just before its first overt unethical step, the monitor predicts whether the trajectory will turn unethical; paired ethical and unethical trajectories per task; default generation parameters; higher is better; the daily life domain. Source: arxiv.org. Saturation forecast: Around 2030. 18 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.4 | 77.9 |
| 2 | Gemma 3 27B | 71 |
| 3 | Qwen 3 0.6B (Non-reasoning) | 66.7 |
| 4 | Gemma 3 4B | 66.4 |
| 5 | GPT-4o | 62.1 |
| 6 | Qwen 3 30B A3B 2507 Instruct | 58 |
| 7 | Qwen 3 0.6B (Thinking) | 54.5 |
| 8 | Qwen 3 8B (Non-reasoning) | 50.7 |
| 9 | Qwen 3 8B (Thinking) | 39.4 |
| 10 | Qwen 3 30B A3B 2507 (Thinking) | 33.2 |
| 11 | Qwen 3 4B 2507 Instruct | 26.1 |
| 12 | Qwen 3 4B 2507 (Thinking) | 20.6 |
| 13 | GPT-OSS-20B | 14.8 |
| 14 | GPT-OSS-120B | 14.7 |
Interactive version: theaggregate.ai/benchmark?slug=preact-bench-daily-life · How It Works · Data refreshed daily, snapshot 2026-09-29.