Pyyan / Open-Weight Models / Qwen3.8-Flash-Next

Qwen3.8-Flash-Next

The most downloaded thing on Hugging Face right now, and a stated Qwen4 preview.

currentAlibabaVerified 2026-09-02

125B

total, 6B active

Specifications

Context262K native, ~1M extendedsource · as of 2026-09-02
Input ($/Mtok)$0.15source · as of 2026-09-02
Output ($/Mtok)$0.47source · as of 2026-09-02
LicenceOpen weightssource · as of 2026-09-02
Max output32Ksource · as of 2026-09-02 · low confidence
Parameters125B total, 6B activesource · as of 2026-09-02
Released26 Aug 2026source · as of 2026-09-02
Model IDQwen/Qwen3.8-Flash-Nextsource · as of 2026-09-02

What matters

  • 125B total parameters with only 6B active per token
  • Alibaba describes it as a preview of the Qwen4 architecture
  • Top of the Hugging Face trending list, with a GGUF build close behind it

Best for open weights at speed.