Safety content moderation on GUARDZOO in-domain (test) benchmark leaderboard · SOTA2 Research

Safety content moderation on GUARDZOO in-domain (test)

89.21Safety ACC (Safe/Unsafe)ROUTEGUARD-4B

Evaluation Results

Method	Links
ROUTEGUARD-4B2026.05		89.21	91.45	70.24	71.88
ROUTEGUARD-8B2026.05		89.19	91.48	70.99	72.21
QWEN3GUARD-4B2026.05		89.17	91.38	69.93	71.27
QWEN3GUARD-8B2026.05		88.99	91.26	70.54	71.84
SHIELDGEMMA-9B2026.05		88.58	90.88	69.66	71.09
LLAMAGUARD2-8B2026.05		88.29	90.6	68.99	70.77
LLAMAGUARD3-8B2026.05		88.2	90.46	68.8	70.35
SHIELDGEMMA-2B2026.05		88.11	90.54	69.37	71.2
LLAMAGUARD3-1B2026.05		87.6	89.94	67.63	70.01
QWEN3GUARD-0.6B2026.05		87.41	89.81	62.65	69.83
ROUTEGUARD-0.6B2026.05		86.28	88.67	67.76	70.58
GRAPH ROUTER2026.05		80.06	82.23	58.66	66.74
DC ROUTER2026.05		79.18	81.34	58.96	66.08
MLP ROUTER2026.05		78.45	80.39	57.84	65
QWEN3GUARD-4B2026.05		77	79.04	21.47	31.11
QWEN3GUARD-8B2026.05		76.14	80.23	26.63	37.5
KNN ROUTER2026.05		71.33	71.45	48.55	55.41
LLAMAGUARD3-8B2026.05		70	69.87	3.86	13.79
SHIELDGEMMA-2B2026.05		68.09	75.13	39.89	48.51
LLAMAGUARD2-8B2026.05		67.71	65.52	2.34	3.56
QWEN3GUARD-0.6B2026.05		66.5	68.83	10.31	23.4
SHIELDGEMMA-9B2026.05		63.74	75.61	40.79	48.38
SHIELDGEMMA-27B2026.05		52.07	55.3	16.72	20.42
LLAMAGUARD3-1B2026.05		45.13	25.2	1.17	3.55