Qwen/Qwen3Guard-Gen-0.6B

Qwen3Guard-Gen-0.6B is a 0.8 billion parameter safety moderation model developed by Qwen, built upon the Qwen3 architecture. This generative model frames safety classification as an instruction-following task, offering three-tiered severity classification (safe, controversial, unsafe) and supporting 119 languages. It excels at both prompt and response classification, providing robust performance for global and cross-lingual safety applications.