ainewsblitz.com

Breaking

Qwen3.8-27B Runs at About 115 Tokens per Second on a Gaming PC

  • Foundation Models
  • Open Source

Alibaba’s newly released Qwen3.8-27B is being run locally on consumer gaming PCs at about 115 tokens per second, a decode rate that testers say puts a 27-billion-parameter open-weight model in interactive territory on a single high-end GPU.

Continue reading

The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.

$20
Read this article
$29/month
Unlimited — all 27,589 articles, the full archive, and comprehension quizzes
Save 72%
$98/year
≈ $8.17/month
Unlimited, billed once a year