Content Moderation on Aegis 2.0
86.1F1 ScoreQwen3Guard-4B-Gen-strict
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen3Guard-4B-Gen-strictStreaming=no, Moderation Policy=Gen-strict, Backbone Scale=4B2026.04 | 86.1 | |
| Qwen3Guard-8B-Gen-strictStreaming=no, Moderation Policy=Gen-strict, Backbone Scale=8B2026.04 | 86.1 | |
| Llama3-StreamGuard-8BStreaming=yes, Backbone Scale=8B2026.04 | 84.5 | |
| Qwen3Guard-Stream-4B-looseStreaming=yes, Moderation Policy=loose, Backbone Scale=4B2026.04 | 83.3 | |
| Llama3-StreamGuard-3BStreaming=yes, Backbone Scale=3B2026.04 | 83.2 | |
| Qwen3Guard-Stream-4B-strictStreaming=yes, Moderation Policy=strict, Backbone Scale=4B2026.04 | 83.1 | |
| Llama3-StreamGuard-1BStreaming=yes, Backbone Scale=1B2026.04 | 82.9 | |
| Qwen3Guard-Stream-8B-strictStreaming=yes, Moderation Policy=strict, Backbone Scale=8B2026.04 | 82.6 | |
| Qwen3Guard-Stream-8B-looseStreaming=yes, Moderation Policy=loose, Backbone Scale=8B2026.04 | 82.4 | |
| Qwen3Guard-Stream-0.6B-strictStreaming=yes, Moderation Policy=strict, Backbone Scale=0.6B2026.04 | 81.4 | |
| Qwen3Guard-Stream-0.6B-looseStreaming=yes, Moderation Policy=loose, Backbone Scale=0.6B2026.04 | 81.4 | |
| MCESetting=Offline2026.01 | 80 | |
| GEPASetting=Offline2026.01 | 76 | |
| ACESetting=Offline2026.01 | 68 | |
| MCE (w/o skills)Setting=Online2026.01 | 68 | |
| MCESetting=Online2026.01 | 63 | |
| ICLSetting=Offline2026.01 | 59 | |
| MIPROv2Setting=Offline2026.01 | 59 | |
| ACESetting=Online2026.01 | 57 | |
| Base ModelSetting=Base2026.01 | 54 | |
| DCSetting=Online2026.01 | 53 |