Medical Advice on Human-Authored
17.51Bypass RateGPT o4-mini
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT o4-mini2026.05 | 17.51 | |
| Llama 4 Scout2026.05 | 20.28 | |
| Gemini 2.5 flash2026.05 | 24.88 | |
| Claude 4.5 Haiku2026.05 | 31.8 | |
| Shieldgemma-9bSafetyguard model=Shieldgemma-9b2026.05 | 32.19 | |
| Qwen3Guard-Gen-8BSafetyguard model=Qwen3Guard-Gen-8B2026.05 | 60.27 | |
| Llama-Guard-3-8BSafetyguard model=Llama-Guard-3-8B2026.05 | 67.1 |