Pyyan / Open-Weight Models / Qwen3.8-27B

Qwen3.8-27B

27B dense under Apache 2.0, and nearly five million downloads.

currentAlibabaVerified 2026-09-02

27B

dense, Apache 2.0

Specifications

Context262K native, ~1M extendedsource · as of 2026-09-02
Input ($/Mtok)Weights onlysource · as of 2026-09-02
Output ($/Mtok)Weights onlysource · as of 2026-09-02
LicenceApache 2.0source · as of 2026-09-02
Max output32Ksource · as of 2026-09-02 · low confidence
Parameters27B densesource · as of 2026-09-02
Released5 Aug 2026source · as of 2026-09-02 · medium confidence
Model IDQwen/Qwen3.8-27Bsource · as of 2026-09-02

What matters

  • 27B dense rather than a mixture of experts, so it behaves predictably on one machine
  • Apache 2.0, weights only, no hosted price
  • One of the most downloaded models on Hugging Face, with quantised builds everywhere

Best for running locally.