HiL-Bench (Public Set Only): leaderboard
Scale HiL-Bench: pass@3 on the200public tasks, separate from its300public/private combined cohort. Exact source version and public-only labels retained with each released model.
Metric: Public Pass@3 (%). Source: labs.scale.com. Saturation forecast: Around February 2027. 3 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Fable 5.2 | 80 | #1 |
| 2 | Claude Fable 5.1 | 61.5 | #14 |
| 3 | Grok 4.20 | 20 | #296 |
No result here: #2 Gemini 4 Carbon, #5 Claude Opus 5.5, #6 GPT-6 Astra.
Interactive version: theaggregate.ai/benchmark?slug=hil-bench-public-set-only · How It Works · Data refreshed daily, snapshot 2026-11-12.