HELM Safety - Simple Safety Tests: leaderboard
The "Simple Safety Tests" subtask of HELM Safety. HELM Safety: Measures model robustness, truthfulness, calibration, bias, harmfulness, jailbreak resistance, or alignment-relevant behavior.
Metric: LM Evaluated Safety score. Source: crfm.stanford.edu. Source: storage.googleapis.com. Saturation forecast: Around February 2027. 87 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Sonnet 4.5 | 100 | #144 |
| 2 | GPT-4.1 Mini | 100 | #358 |
| 3 | Qwen 2.5 72B Instruct | 100 | #463 |
| 4 | GPT-OSS-120B | 100 | #344 |
| 5 | GPT-5 Nano | 100 | #430 |
| 6 | Claude 3.5 Sonnet (20240620) | 100 | #300 |
| 7 | GPT-OSS-20B | 100 | #526 |
| 8 | GPT-4.1 (2025-04-14) | 100 | #252 |
| 9 | Kimi K2 | 100 | #244 |
| 10 | GPT-5 Mini (2025-08-07) | 100 | #175 |
| 11 | O4 Mini (2025-04-16) | 100 | #182 |
| 12 | Claude 3.7 Sonnet (20250219) | 100 | #205 |
| 13 | Claude 3 Haiku (20240307) | 100 | #783 |
| 14 | Claude 3 Sonnet (20240229) | 100 | #686 |
| 15 | Claude 3 Opus (20240229) | 100 | #420 |
No result here: #4 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=helm-safety-simple-safety-tests · How It Works · Data refreshed daily, snapshot 2026-10-24.