Prompt Classification on ToxicChat Text Prompt
96.27F1 ScoreGPT4o-mini
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT4o-mini2025.12 | 96.27 | |
| ProGuard-7B2025.12 | 96.07 | |
| GuardReasoner-8B2025.12 | 95.8 | |
| Gemini2.5-Flash2025.12 | 95.43 | |
| GuardReasonerVL-7B2025.12 | 95.32 | |
| ProGuard-3B2025.12 | 95.03 | |
| LlamaGuard-7B2025.12 | 94.53 | |
| ShieldGemma-9B2025.12 | 93.97 | |
| GPT-OSS-SafeGuard-20B2025.12 | 93.72 | |
| LlamaGuard2-8B2025.12 | 92.33 | |
| LlamaGuard3-11B-Vision2025.12 | 92.05 | |
| LlamaGuard3-8B2025.12 | 91.56 | |
| LlamaGuard4-12B2025.12 | 91.15 | |
| WildGuard-7B2025.12 | 88.99 |