Response Classification on EXPGUARD (test)
96.7Financial ScoreEXPGUARD
Evaluation Results
| Method | Links | ||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| EXPGUARDModel Size=7B2026.03 | 96.7 | 86.2 | 92.4 | 92.7 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Aegis-Guard-DModel Size=7B2026.03 | 91.8 | 78 | 88.1 | 87.2 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Llama-Guard3Model Size=8B2026.03 | 87.6 | 81.9 | 80.4 | 84.2 | — | — | — | — | — | — | — | — | — | — | — | — | |
| MD-JudgeModel Size=7B2026.03 | 84.2 | 70.9 | 87.6 | 81.4 | — | — | — | — | — | — | — | — | — | — | — | — | |
| EXPGUARDModel Size=7B2026.03 | 83.3 | 42.7 | 66.7 | 64.2 | — | — | — | — | — | — | — | — | — | — | — | — | |
| WildGuardModel Size=7B2026.03 | 82.4 | 62.8 | 83.1 | 77.4 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Llama-Guard2Model Size=8B2026.03 | 81.7 | 71.2 | 68 | 75.3 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Aegis-Guard-PModel Size=7B2026.03 | 81.2 | 57.8 | 76.2 | 73.9 | — | — | — | — | — | — | — | — | — | — | — | — | |
| MD-JudgeModel Size=7B2026.03 | 78.3 | 33.7 | 58.3 | 56.8 | — | — | — | — | — | — | — | — | — | — | — | — | |
| BeaverDamModel Size=7B2026.03 | 76.9 | 59.2 | 69.1 | 70.3 | — | — | — | — | — | — | — | — | — | — | — | — | |
| WildGuardModel Size=7B2026.03 | 73.6 | 47.7 | 71.2 | 64.2 | — | — | — | — | — | — | — | — | — | — | — | — | |
| HarmBench-MistralModel Size=7B2026.03 | 70.7 | 63 | 70.5 | 68.5 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Llama-Guard3Model Size=8B2026.03 | 67.3 | 39.7 | 37.2 | 48.1 | — | — | — | — | — | — | — | — | — | — | — | — | |
| HarmBench-LlamaModel Size=13B2026.03 | 66.2 | 59.6 | 67.7 | 64.5 | — | — | — | — | — | — | — | — | — | — | — | — | |
| HarmBench-LlamaModel Size=13B2026.03 | 61.3 | 57.2 | 46.8 | 55.1 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Llama-GuardModel Size=7B2026.03 | 58.8 | 23.7 | 44.8 | 46.5 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Llama-Guard1Model Size=7B2026.03 | 56.9 | 12.1 | 35.9 | 35 | — | — | — | — | — | — | — | — | — | — | — | — | |
| BeaverDamModel Size=7B2026.03 | 54.4 | 28.6 | 35.9 | 39.6 | — | — | — | — | — | — | — | — | — | — | — | — | |
| ShieldGemmaModel Size=9B2026.03 | 49.4 | 19.1 | 40.4 | 36.3 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Aegis-Guard-PModel Size=7B2026.03 | 31.1 | 19.6 | 54.5 | 35.1 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Aegis-Guard-DModel Size=7B2026.03 | 30.8 | 18.1 | 53.2 | 34 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Llama-Guard2Model Size=8B2026.03 | 26.4 | 11.1 | 4.5 | 14 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Llama-Guard2Model Size=8B2026.03 | 25.5 | 39.6 | 33.9 | 33 | — | — | — | — | — | — | — | — | — | — | — | — | |
| HarmBench-MistralModel Size=7B2026.03 | 17.2 | 12.3 | 17.7 | 15.7 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Aegis-Guard-DModel Size=7B2026.03 | 8.3 | 7 | 12.9 | 9.4 | — | — | — | — | — | — | — | — | — | — | — | — | |
| HarmBench-MistralModel Size=7B2026.03 | 7.9 | 0.5 | 5.8 | 4.7 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Llama-Guard3Model Size=8B2026.03 | 4.4 | 13.9 | 9.7 | 9.3 | — | — | — | — | — | — | — | — | — | — | — | — | |
| EXPGUARDModel Size=7B2026.03 | 3.4 | 6.4 | 10.5 | 6.8 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Aegis-Guard-PModel Size=7B2026.03 | 2 | 1.1 | 0 | 1 | — | — | — | — | — | — | — | — | — | — | — | — | |
| BeaverDamModel Size=7B2026.03 | 2 | 1.6 | 4 | 2.5 | — | — | — | — | — | — | — | — | — | — | — | — | |
| ShieldGemmaModel Size=9B2026.03 | 1.3 | 20.7 | 9.8 | 9.3 | — | — | — | — | — | — | — | — | — | — | — | — | |
| WildGuardModel Size=7B2026.03 | 1 | 0 | 2.4 | 1.1 | — | — | — | — | — | — | — | — | — | — | — | — | |
| AzureModel Size=-2026.03 | 0.6 | 4.9 | 3.8 | 2.6 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Llama-Guard1Model Size=7B2026.03 | 0.5 | 1.1 | 0 | 0.5 | — | — | — | — | — | — | — | — | — | — | — | — | |
| MD-JudgeModel Size=7B2026.03 | 0.5 | 2.7 | 2.4 | 1.9 | — | — | — | — | — | — | — | — | — | — | — | — | |
| ShieldGemmaModel Size=9B2026.03 | 0 | 0 | 0 | 0 | — | — | — | — | — | — | — | — | — | — | — | — | |
| DetoxifyModel Size=-2026.03 | 0 | 2 | 0 | 0.6 | — | — | — | — | — | — | — | — | — | — | — | — | |
| PerspectiveModel Size=-2026.03 | 0 | 2 | 0 | 0.6 | — | — | — | — | — | — | — | — | — | — | — | — | |
| OpenAI ModerationModel Size=-2026.03 | 0 | 2 | 0 | 0.6 | — | — | — | — | — | — | — | — | — | — | — | — | |
| HarmBench-LlamaModel Size=13B2026.03 | 0 | 1.5 | 0.6 | 0.7 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Aegis-Guard-DModel Size=7B2026.03 | — | — | — | — | 14.2 | 41.7 | 10.3 | 22 | — | — | — | — | — | — | — | — | |
| Aegis-Guard-DModel Size=7B2026.03 | — | — | — | — | — | — | — | — | 77.7 | 54.3 | 76.3 | 69.4 | — | — | — | — | |
| Aegis-Guard-DModel Size=7B2026.03 | — | — | — | — | — | — | — | — | — | — | — | — | 22 | 7.1 | 22 | 17.1 | |
| Aegis-Guard-PModel Size=7B2026.03 | — | — | — | — | 62.3 | 76.4 | 57.7 | 65.4 | — | — | — | — | — | — | — | — | |
| Aegis-Guard-PModel Size=7B2026.03 | — | — | — | — | — | — | — | — | 79.9 | 57.8 | 70.5 | 69.4 | — | — | — | — | |
| Aegis-Guard-PModel Size=7B2026.03 | — | — | — | — | — | — | — | — | — | — | — | — | 12.1 | 17 | 7.7 | 12.2 | |
| BeaverDamModel Size=7B2026.03 | — | — | — | — | 36.8 | 57.3 | 45.5 | 46.5 | — | — | — | — | — | — | — | — | |
| BeaverDamModel Size=7B2026.03 | — | — | — | — | — | — | — | — | 73.6 | 57.8 | 65.4 | 65.6 | — | — | — | — | |
| BeaverDamModel Size=7B2026.03 | — | — | — | — | — | — | — | — | — | — | — | — | 3.7 | 12.5 | 8 | 8 | |
| EXPGUARDModel Size=7B2026.03 | — | — | — | — | 4.4 | 20.1 | 7.1 | 10.5 | — | — | — | — | — | — | — | — | |
| EXPGUARDModel Size=7B2026.03 | — | — | — | — | — | — | — | — | 97.2 | 74.9 | 87.8 | 86.6 | — | — | — | — | |
| EXPGUARDModel Size=7B2026.03 | — | — | — | — | — | — | — | — | — | — | — | — | 1.7 | 5.7 | 3.9 | 3.8 | |
| HarmBench-LlamaModel Size=13B2026.03 | — | — | — | — | 31.4 | 34.7 | 29.5 | 31.9 | — | — | — | — | — | — | — | — | |
| HarmBench-LlamaModel Size=13B2026.03 | — | — | — | — | — | — | — | — | 0.9 | 4 | 1.3 | 2.1 | — | — | — | — | |
| HarmBench-LlamaModel Size=13B2026.03 | — | — | — | — | — | — | — | — | — | — | — | — | 36.5 | 40.1 | 30.7 | 35.8 | |
| HarmBench-MistralModel Size=7B2026.03 | — | — | — | — | 39.3 | 48.7 | 37.8 | 42 | — | — | — | — | — | — | — | — | |
| HarmBench-MistralModel Size=7B2026.03 | — | — | — | — | — | — | — | — | 13.2 | 39.7 | 44.2 | 32.4 | — | — | — | — | |
| HarmBench-MistralModel Size=7B2026.03 | — | — | — | — | — | — | — | — | — | — | — | — | 19.9 | 22.7 | 19.3 | 20.6 | |
| Llama-Guard1Model Size=7B2026.03 | — | — | — | — | 60.4 | 86.4 | 71.2 | 72.7 | — | — | — | — | — | — | — | — | |
| Llama-Guard1Model Size=7B2026.03 | — | — | — | — | — | — | — | — | 78 | 55.8 | 68.6 | 67.5 | — | — | — | — | |
| Llama-Guard1Model Size=7B2026.03 | — | — | — | — | — | — | — | — | — | — | — | — | 23.3 | 37.4 | 27.2 | 29.3 | |
| Llama-Guard2Model Size=8B2026.03 | — | — | — | — | 19.5 | 23.6 | 35.3 | 26.1 | — | — | — | — | — | — | — | — | |
| Llama-Guard2Model Size=8B2026.03 | — | — | — | — | — | — | — | — | 38.7 | 23.6 | 10.9 | 24.4 | — | — | — | — | |
| Llama-Guard2Model Size=8B2026.03 | — | — | — | — | — | — | — | — | — | — | — | — | 9.2 | 19.1 | 19.5 | 15.9 | |
| Llama-Guard3Model Size=8B2026.03 | — | — | — | — | 20.1 | 22.6 | 28.8 | 23.9 | — | — | — | — | — | — | — | — | |
| Llama-Guard3Model Size=8B2026.03 | — | — | — | — | — | — | — | — | 81.8 | 51.3 | 62.2 | 65.1 | — | — | — | — | |
| Llama-Guard3Model Size=8B2026.03 | — | — | — | — | — | — | — | — | — | — | — | — | 6.5 | 6.7 | 8.9 | 7.4 | |
| MD-JudgeModel Size=7B2026.03 | — | — | — | — | 27.4 | 43.7 | 20.5 | 30.5 | — | — | — | — | — | — | — | — | |
| MD-JudgeModel Size=7B2026.03 | — | — | — | — | — | — | — | — | 92.1 | 68.8 | 87.8 | 82.9 | — | — | — | — | |
| MD-JudgeModel Size=7B2026.03 | — | — | — | — | — | — | — | — | — | — | — | — | 3.4 | 12.5 | 2.9 | 6.3 | |
| ShieldGemmaModel Size=9B2026.03 | — | — | — | — | 99.4 | 88.9 | 94.9 | 94.4 | — | — | — | — | — | — | — | — | |
| ShieldGemmaModel Size=9B2026.03 | — | — | — | — | — | — | — | — | 84.3 | 41.2 | 50 | 58.5 | — | — | — | — | |
| ShieldGemmaModel Size=9B2026.03 | — | — | — | — | — | — | — | — | — | — | — | — | 58.7 | 38.4 | 49.1 | 48.7 | |
| WildGuardModel Size=7B2026.03 | — | — | — | — | 29.6 | 54.3 | 27.6 | 37.1 | — | — | — | — | — | — | — | — | |
| WildGuardModel Size=7B2026.03 | — | — | — | — | — | — | — | — | 95.6 | 66.3 | 84 | 82 | — | — | — | — | |
| WildGuardModel Size=7B2026.03 | — | — | — | — | — | — | — | — | — | — | — | — | 12.3 | 23.2 | 12.5 | 16 |