Experiment with Qwen3.8-Flash-Next 176B Model on NVIDIA GB300 NVL72 for Agentic Coding
Alibaba previews Qwen4 architecture via 176B MoE model with 1M token context
Alibaba released preview weights for Qwen3.8-Flash-Next, a 176B multimodal MoE model that activates only 6B parameters per token, offering a first look at the upcoming Qwen4 architecture. The model features a native 262K-token context window extensible to 1M tokens via YaRN, positioned for agentic coding workloads on NVIDIA GB300 NVL72 hardware. This signals Alibaba's continued push into frontier open-weight models competitive with leading closed models.