Qwen3-TTS

Qwen3 is a series of large language models developed by the Qwen team at Alibaba Cloud. It offers a range of models, including dense and Mixture-of-Expert (MoE) architectures, with various sizes from 0.6B to 235B-A22B parameters. The models are designed for diverse applications, featuring strong capabilities in instruction following, logical reasoning, text comprehension, mathematics, science, coding, and tool usage across multiple languages. The Qwen3 series includes specialized variants like Qwen3-Instruct-2507 for general capabilities and Qwen3-Thinking-2507 for enhanced reasoning tasks. It supports long-context understanding up to 1 million tokens and excels in human preference alignment for creative writing, role-playing, and multi-turn dialogues. The models are suitable for developers and researchers looking to integrate advanced AI into their applications, offering guidance for inference, local execution, deployment, quantization, and post-training with popular frameworks.