SCOUT-450 (LLM Judge): leaderboard
Metric: Accuracy (%; binary attack-or-benign verdicts on 450 held-out samples, one query per sample with a shared evaluation-only prompt). Source: arxiv.org. Saturation forecast: Around December 2026. 5 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4o | 92.4 |
| 2 | GPT-5.1 | 87.3 |
| 3 | GPT-5.2 | 80.4 |
| 4 | Gemini 3.1 Pro (Preview) | 66.9 |
Interactive version: theaggregate.ai/benchmark?slug=scout-450-llm-judge · How It Works · Data refreshed daily, snapshot 2026-09-26.