Our new X account is live! Follow @wizwand_team for updates

Text-based safety moderation on OpenAI

82.3F1 Score

GPT-4o

Updated 4d ago

Evaluation Results

Method	Links
GPT-4o 2025.12		82.3	88.6
LLaMA Guard 3 2025.12		81.6	85.7
OMNIGUARD-7B 2025.12		81.1	87.9
Qwen3Guard 2025.12		81	-
gpt-oss-safeguard 2025.12		81	-
Qwen3-235B 2025.12		80.1	85.4
Qwen2.5-72B 2025.12		79.8	85.2
ShieldGemma 2025.12		79	-
Llama Guard 3 2025.12		79	-
ThinkGuard 2025.12		78.7	79
OMNIGUARD-3B 2025.12		77.8	83.8
IBM Granite Guardian 3.3 2025.12		77	-
AprielGuard 2025.12		77	-
IBM Granite Guardian 3.3 2025.12		77	-
Llama Guard 2 2025.12		76	-
AprielGuard 2025.12		75	-
LLaMA Guard 2 2025.12		74.4	85.6
Llama Guard 4 2025.12		73	-
IBM Granite Guardian 3.2 2025.12		73	-
Qwen2.5-7B 2025.12		72.6	81.4
Qwen2.5-Omni-7B 2025.12		70	70.8
IBM Granite Guardian 3.1 2025.12		69	-
IBM Granite Guardian 3.2 2025.12		68	-
Qwen3Guard 2025.12		68	-
LLaMA-3.3-70B 2025.12		58.4	80.5
LLaMA Guard 1 2025.12		32.8	74.4