ainewsblitz.com

Breaking

Qwen releases open-weight Qwen3.8-27B with 262K native context and local GPU support

Qwen releases open-weight Qwen3.8-27B with 262K native context and local GPU support
  • Foundation Models
  • Software Dev & Coding
  • Open Source

Qwen has released Qwen3.8-27B, a 27-billion-parameter open-weight vision-language model aimed at coding, professional work and long-running agent tasks. A highly optimized local demonstration reported generation at 206 tokens per second on one NVIDIA GeForce RTX 5090, although that result depends on aggressive quantization and speculative decoding rather than representing standard performance.

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 27,609 articles, the full archive, and comprehension quizzes
Save 72%
$98/year
≈ $8.17/month
Unlimited, billed once a year