Prompt Classification on Simple SafetyTest Text Prompt
100F1 ScoreGemini2.5-Flash
Evaluation results
| Method | Links | |
|---|---|---|
| Gemini2.5-Flash2025.12 | 100 | |
| GPT4o-mini2025.12 | 100 | |
| GuardReasoner-8B2025.12 | 99.5 | |
| GuardReasonerVL-7B2025.12 | 99.5 | |
| ProGuard-3B2025.12 | 99.5 | |
| ProGuard-7B2025.12 | 99.5 | |
| LlamaGuard3-8B2025.12 | 98.99 | |
| LlamaGuard4-12B2025.12 | 98.48 | |
| LlamaGuard3-11B-Vision2025.12 | 97.96 | |
| GPT-OSS-SafeGuard-20B2025.12 | 97.5 | |
| LlamaGuard2-8B2025.12 | 95.83 | |
| LlamaGuard-7B2025.12 | 92.47 | |
| WildGuard-7B2025.12 | 90.71 | |
| ShieldGemma-9B2025.12 | 85.06 |