What is Qwen and what is it used for?
Qwen is the family of language and multimodal models developed by the Qwen team at Alibaba Cloud, with open weights published on Hugging Face and ModelScope and API access through Alibaba Cloud Model Studio.
The official repository describes the series as language models from the Qwen team at Alibaba Cloud. Several models have open weights to download and run on your own infrastructure. With the launch of the Qwen3 series, the team states that all its open-weight models use the Apache 2.0 license.
The same family covers different tasks:
- General-purpose language models, with versions that reason before answering.
- Qwen-VL, which reads images and video together with text.
- Qwen-Omni, which combines text, image, audio and video.
- Qwen-Coder, for code.
- Embedding and reranking models, for search.
- Qwen-Image, for images.
Through the Model Studio API, the documentation describes an interface compatible with the OpenAI chat format and lists models in a commercial version and an open version.
