Qwen3.8-Flash-Next activates 6B of its 125B parameters per token. Its licence, not its architecture, is the part that matters under the EU AI Act.

Alibaba's Qwen team is teasing its next architecture a day early—and the specs say it runs near-frontier scale on a fraction of the power.

Alibaba’s Qwen team is scheduled to open-source Qwen3.8-Flash-Next and an FP8 version at 11 p.m. Beijing time on Aug. 26. The model is described as a