Story thread · 2 reports / 2 sources

Trying out local AI: This is what Qwen3.8-27B can do

heise.de · 22h · first report

How the coverage leans

Across 2 sources · syndicated copies counted once

Alibaba has released the weights of its Qwen3.8-27B model, small enough for affordable graphics cards. The gap to top models is shrinking.

The coverage

  1. Qwen3.8-27B at 256K on a 24GB RTX PRO 4000 SFF (432 GB/s): 50 tok/s with MTP

    news.ycombinator.com · 3h

The conversation · 0

Sign in to join the conversation.

No comments yet — start the thread.