Qwen3-1.7B is the latest generation of large language models in the Qwen series, offering advanced capabilities in reasoning, instruction-following, agent functionality, and multilingual support. A distinctive feature is its unique ability to seamlessly switch between thinking mode for complex logical reasoning, mathematics, and coding tasks, and non-thinking mode for efficient general-purpose dialogue, all within a single model. The model significantly enhances reasoning capabilities beyond previous versions like QwQ and Qwen2.5, with superior human preference alignment for creative writing, role-playing, multi-turn dialogues, and instruction following. With 1.7 billion parameters across 28 layers, a context length of 32,768 tokens, and support for over 100 languages and dialects, Qwen3-1.7B delivers strong multilingual instruction-following and translation capabilities. This MLX-4bit version is optimized for efficient inference on Apple Silicon hardware.