Alibaba Tongyi Qwen3.8-Flash released as open source: 125B parameter MoE model with training costs just one-ninth of the previous generation
This new model features a 125B parameter MoE architecture, natively supporting a 260k token context window, expandable to 1M. It will soon launch an open API on the Qwen AI platform, and the model weights have been simultaneously open-sourced.
Return to original text
A short introduction is retained here; the full text from the source website is not reposted. For specific wording and subsequent updates, please refer to the original article.
Open the original article on ITHome