Qwen is a research team dedicated to developing AGI models that will shape the future of humanity. Our mission is to create omni and smart models that are safe and responsible, while making AGI accessible to the global community through open-source initiatives. We have developed a diverse portfolio of AGI models, including our flagship large language model series: Qwen2.5 for general language tasks, QwQ for enhanced reasoning capabilities, Qwen2.5-Coder for coding development, and Qwen2.5-Math for mathematical computation. Our multimodal models include Qwen2-VL for vision-language understanding and Qwen2-Audio for audio understanding. Additionally, we have developed the Qwen-Agent framework to transform these models into versatile agents capable of more sophisticated applications. Qwen Chat serves as an intuitive interface enabling users and developers to seamlessly interact with both our open-source and proprietary models, including Qwen2.5-Plus, QwQ, and Qwen2-VL-Max. The platform offers comprehensive functionality spanning image and video understanding, image generation, document processing, web search integration, tool utilization, and artifacts.