Qwen3-0.6B

Qwen3-0.6B is a 0.6-billion-parameter open-weight text-generation model from Alibaba's Qwen team, designed for lightweight chat and instruction-following with support for thinking/non-thinking dual modes, 32K context, and 100+ languages. It is the smallest dense model in the Qwen3 family, optimized for edge and resource-constrained deployments. Source: supabase.