This 4B Safety Model Classifies AI Content as Safe, Unsafe, or Controversial

qwen3guard-gen-4b is a compact 4B moderation model that classifies content across 119 languages as Safe, Unsafe, or Controversial.