ClawsBench - Safety (OpenClaw; skills off; meta on): leaderboard
Metric: Mean native task score x100 (-100 to 100); OpenClaw, domain skills off, meta prompt on; mean over observed scored trials, including scored error rows; higher is better. Source: huggingface.co. 3 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 37.37 |
| 2 | Claude Sonnet 4.6 | 30.76 |
| 3 | Gemini 3.1 Flash Lite (Preview) | 0.88 |
Interactive version: theaggregate.ai/benchmark?slug=clawsbench-safety-openclaw-skills-off-meta-on · How It Works · Data refreshed daily, snapshot 2026-10-09.