Response Classification on Wild Guard Text Response
93.17F1 ScoreDynaGuard-8B
Evaluation results
| Method | Links | |
|---|---|---|
| DynaGuard-8B2025.12 | 93.17 | |
| GuardReasonerVL-7B2025.12 | 93.07 | |
| GuardReasoner-8B2025.12 | 93.04 | |
| ProGuard-7B2025.12 | 92.92 | |
| MD-Judge-7B2025.12 | 91.36 | |
| LlamaGuard3-8B2025.12 | 90.29 | |
| LlamaGuard2-8B2025.12 | 89.87 | |
| LlamaGuard4-12B2025.12 | 89.87 | |
| ProGuard-3B2025.12 | 89.36 | |
| Gemini2.5-Flash2025.12 | 89.08 | |
| LlamaGuard3-11B-Vision2025.12 | 87.19 | |
| LlamaGuard-7B2025.12 | 84.99 | |
| GPT-OSS-SafeGuard-20B2025.12 | 80.45 | |
| WildGuard-7B2025.12 | 80.16 | |
| ShieldGemma-9B2025.12 | 78.88 | |
| GPT4o-mini2025.12 | 78.83 |