Qwen is a series of large language and multimodal models developed by the Qwen Team at Alibaba Cloud.
The Qwen 2.5 family was released in September 2024. It quickly became a formidable competitor in the open-source LLM space, performing well in all tasks, such as processing structured data, generating structured outputs, computer coding, and multilingual support. These models range from 0.5b to 72b, providing a comprehensive range for nearly all regular use cases.
Qwen Models & Services
Qwen2.5-Coder Series
Qwen2.5-Coder series, released on 12th November 2024, achieved “the best performance among open-source models on multiple popular code generation benchmarks” [@qwenteamQwen25CoderSeriesPowerful2024]. Key features include code generation, code repair and code reasoning. They are expected to be available on the Tongyi official website, supporting the one-click generation of websites, mini-games, and data charts, among other visual applications.
References
- Mistral AI Team (2024) Mistral Moderation API. https://mistral.ai/news/mistral-moderation/.
- Qwen Team (2024) Qwen2.5-Coder Series: Powerful, Diverse, Practical. Qwen. http://qwenlm.github.io/blog/qwen2.5-coder-family/.