Prompt Classification on OpenAI Moderation Text Prompt
88.89F1 ScoreLlamaGuard2-8B
Evaluation Results
| Method | Links | |
|---|---|---|
| LlamaGuard2-8B2025.12 | 88.89 | |
| GPT4o-mini2025.12 | 88.64 | |
| GPT-OSS-SafeGuard-20B2025.12 | 87.42 | |
| LlamaGuard3-8B2025.12 | 86.5 | |
| Gemini2.5-Flash2025.12 | 86.1 | |
| LlamaGuard-7B2025.12 | 85.05 | |
| LlamaGuard4-12B2025.12 | 84.83 | |
| ProGuard-7B2025.12 | 83.05 | |
| LlamaGuard3-11B-Vision2025.12 | 82.71 | |
| GuardReasonerVL-7B2025.12 | 78.74 | |
| ShieldGemma-9B2025.12 | 78.69 | |
| GuardReasoner-8B2025.12 | 77.67 | |
| ProGuard-3B2025.12 | 77.02 | |
| WildGuard-7B2025.12 | 70.43 |