List your company

Response Classification on BeaverTails V Text-Image Response

84.8F1 ScoreQwen3Guard-8B-Gen
55.88863.39470.978.406Dec 29, 2025Jan 5, 2026Jan 12, 2026Jan 19, 2026Jan 26, 2026Feb 2, 2026Feb 10, 2026

Evaluation results

MethodLinks
Qwen3Guard-8B-GenSafeguard Type=Post-hoc2026.0284.8
Qwen3Guard-4B-StreamSafeguard Type=Streaming2026.0284.6
Qwen3Guard-4B-GenSafeguard Type=Post-hoc2026.0284.5
Qwen3Guard-8B-StreamSafeguard Type=Streaming2026.0284.2
KelpSafeguard Type=Streaming2026.0284.1
GuardReasoner-8B2025.1284.02
Qwen3Guard-0.6B-GenSafeguard Type=Post-hoc2026.0284
SCM-7BSafeguard Type=Streaming2026.0283.4
ProGuard-7B2025.1283.1
WildGuard-7BSafeguard Type=Post-hoc2026.0282.9
Qwen3Guard-0.6B-StreamSafeguard Type=Streaming2026.0282.7
GuardReasonerVL-7B2025.1282.35
DynaGuard-8B2025.1281.73
NExT-GuardSafeguard Type=Streaming2026.0281.2
MD-Judge-7B2025.1281.13
Gemini2.5-Flash2025.1280.72
NExT-Guard*Safeguard Type=Streaming2026.0280.6
ProGuard-7B2025.1280.44
ProGuard-3B2025.1279.83
ProGuard-3B2025.1279.19
Gemini2.5-Flash2025.1277.85
NemoGuard-8BSafeguard Type=Post-hoc2026.0277.4
LlamaGuard2-8B2025.1274.62
LlamaGuard4-12B2025.1273.51
WildGuard-7B2025.1273.39
GPT4o-mini2025.1272.92
GPT-OSS-SafeGuard-20B2025.1272.59
LlamaGuard3-8B2025.1272.4
LlamaGuard3-11B-Vision2025.1270.91
LlamaGuard-7B2025.1269.49
LlamaGuard4-12BSafeguard Type=Post-hoc2026.0267.5
LlamaGuard3-8BSafeguard Type=Post-hoc2026.0267.3
ShieldGemma-27BSafeguard Type=Post-hoc2026.0266.8
LlamaGuard3-11B-Vision2025.1266.62
GuardReasonerVL-7B2025.1266.13
GPT4o-mini2025.1263.5
LlamaGuard4-12B2025.1262.8
ShieldGemma-9BSafeguard Type=Post-hoc2026.0261.1
ShieldGemma-9B2025.1257