Prompt Classification on XSTest Text Prompt
93.71F1 ScoreGuardReasoner-8B
Evaluation Results
| Method | Links | |
|---|---|---|
| GuardReasoner-8B2025.12 | 93.71 | |
| Gemini2.5-Flash2025.12 | 90.91 | |
| GPT-OSS-SafeGuard-20B2025.12 | 90.81 | |
| GuardReasonerVL-7B2025.12 | 90.59 | |
| LlamaGuard2-8B2025.12 | 90.57 | |
| LlamaGuard3-8B2025.12 | 89.98 | |
| ProGuard-3B2025.12 | 88.25 | |
| ProGuard-7B2025.12 | 87.9 | |
| LlamaGuard3-11B-Vision2025.12 | 87.19 | |
| LlamaGuard4-12B2025.12 | 86.64 | |
| GPT4o-mini2025.12 | 85.98 | |
| LlamaGuard-7B2025.12 | 83.42 | |
| WildGuard-7B2025.12 | 81.91 | |
| ShieldGemma-9B2025.12 | 78.81 |