Enkrypt AI - Safety Risk: leaderboard
Enkrypt AI’s LLM safety leaderboard: how often the model produces unsafe content on red-team prompts. It is reported as a risk, so lower is better here and the leader is the least risky model rather than the most capable one.
Metric: Risk Score. Source: leaderboard.enkryptai.com. Saturation forecast: Not forecast. 298 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Fable 5 | 0 | #16 |
| 2 | Mistral-NeMo-Minitron-8B-Instruct | 3.67 | #907 |
| 3 | Claude 3 Opus (20240229) | 7.34 | #412 |
| 4 | Claude 3.5 Sonnet (20241022) | 10.67 | #291 |
| 5 | Claude Opus 5.5 | 12.03 | #3 |
| 6 | Claude Sonnet 5 | 13.31 | #57 |
| 7 | Qwen 2.5 72B Instruct Turbo | 13.47 | #527 |
| 8 | Claude Sonnet 4 (20250514) | 14.32 | #213 |
| 9 | GPT-5 Nano | 14.33 | #425 |
| 10 | GPT-5 | 14.39 | #98 |
| 11 | Claude 3.5 Haiku (20241022) | 15.01 | #588 |
| 12 | GPT-4 Preview (0125) | 15.97 | #486 |
| 13 | GPT-5 Mini | 16.02 | #186 |
| 14 | Grok 4.7 | 16.52 | #25 |
| 15 | Claude Opus 4 (20250514) | 16.79 | #154 |
No result here: #6 GPT-6 Astra, #7 GPT-6.1 Sol, #8 Gemini 4 Argon.
Interactive version: theaggregate.ai/benchmark?slug=enkrypt-ai-safety-risk · How It Works · Data refreshed daily, snapshot 2026-10-20.