Qwen3.8-27B is free on Sprag through the end of August
Sprag
2026-08-161 min read
Qwen3.8-27B is open weights under Apache 2.0 and built to be run locally. Running it locally still takes a 24GB card, so we're serving it for free through the end of August.
Qwen3.8-27B is open weights under Apache 2.0 and built to be run locally. Running it locally still takes a 24GB card, and plenty of the people most interested in this model don't have one.
Why we're hosting it for free
So we're hosting it on the Sprag API at no cost through at least the end of August. Generate a key and use it. If the model turns out to be worth the hardware, go buy the card and run it yourself.
Why we care about this
Sprag is built on vLLM-Omni, which we develop in the open. Making models like this one practical to serve is the work we already do, and for the next few weeks we're covering the GPUs.
Getting started
Everything you need is on the model page.