Product LaunchSimon Willison

Qwen3.8-Flash-Next

#qwen#ai#model#multimodal#performance

English

Qwen has released a new multimodal MoE model, Qwen3.8-Flash-Next, which serves as a preview for the upcoming Qwen4 architecture. The model features 125 billion tokens with 6 billion active, providing a significant performance boost, and has been tested using quantized models on a DGX Spark.

中文

Qwen发布了新的多模态MoE模型Qwen3.8-Flash-Next，该模型作为即将推出的Qwen4架构的预览。该模型具有1250亿个标记和60亿个活跃标记，提供显著的性能提升，并已在DGX Spark上使用量化模型进行了测试。