Model
gpt-oss-safeguard-20b
An open-weight safety moderation model with 20B parameters, accessed through OpenRouter, intended for content safety classification tasks.
Timeline (last 90 days)
Oct 13
In a test of 360 clean texts, Jev made only 1 misjudgment, gpt-5-mini made 1 misjudgment, Gemini 2.5 Flash-Lite and Gemma each made 2 misjudgments, gpt-oss-safeguard-20b made 4 misjudgments, and Llama Guard 4 made 26 misjudgments.
Unverified50%