Explore Qwen, an open-source AI model focused on AGI development. Discover powerful language & multimodal models, including Qwen TTS for natural, cost-effective voice generation.
The Qwen team is advancing artificial general intelligence (AGI) through the development of versatile models, including large language and multimodal AI. Their mission centers on creating safe and responsible AGI accessible to all via open-source initiatives. Recent releases include the Qwen3 series of large language models, featuring dense and Mixture-of-Experts architectures, alongside specialized models like Qwen3-Coder for enhanced programming. In the multimodal domain, they offer Qwen-Image (20B parameters) and Qwen-Image-Edit, along with the Qwen-Agent framework for creating intelligent agents utilizing instruction following, tool use, and planning. A key platform is Qwen Studio, offering functionalities like image and video understanding, generation, document processing, web search, and tool utilization. The latest model, Qwen3.7 Plus, integrates advanced text and multimodal processing for tasks such as web development, complex reasoning, role-playing, creative writing, visual reasoning, OCR, and spatial understanding with a 1,000,000 token context length and 65,536 token summary generation capability.
