Qwen发布Qwen3.8-Flash-Next开放权重预览
Qwen发布了后训练版Qwen3.8-Flash-Next的权重和配置文件。这是一款1800亿参数的图文模型，也是Qwen4所采用架构的实验性预览；Qwen称其兼容Hugging Face Transformers、vLLM、SGLang和TokenSpeed，并提供Qwen Cloud托管API。Qwen表示，混合注意力、门控残差和n-gram嵌入等设计可提升长上下文效率与参数扩展能力。
此次发布让开发者能够直接获取Qwen下一代架构的开放权重预览，并通过常用推理框架或Qwen Cloud托管API进行部署。
Qwen/Qwen3.8-Flash-Next Image-Text-to-Text • 180B • Updated 4 days ago • 3.41M • • 6.9k
library_name: transformers license: other license_name: qwen-community-1.0 license_link: LICENSE pipeline_tag: image-text-to-text
Qwen3.8-Flash-Next
[!Note] This repository contains model weights and configuration files for the post-trained model in the Hugging Face Transformers format. These artifacts are compatible with Hugging Face Transformers, vLLM, SGLang, TokenSpeed, etc.
[!Tip] For users seeking managed, scalable inference without infrastructure maintenance, the official Qwen API service is provided by Qwen Cloud. In particular, Qwen3.8-Flash is the official version based on Qwen3.8-Flash-Next with more production features, e.g., 1M context length by default, official built-in tools. For more information, please refer to the Qwen3.8-Flash Overview.
#Qwen#Qwen3.8-Flash-Next#Qwen3.8-Flash#Qwen4#Qwen Cloud#Hugging Face Transformers#vLLM#SGLang#TokenSpeed