Alibaba's Qwen3.8-27B Scores 67.3 Overall, Matching GPT-5.4 After a Jump From 42.2
Sign in✓ Signed in
Alibaba's Qwen team has released Qwen3.8-27B, a 27-billion-parameter open-weight model that a circulating composite overall score places at 67.3 — essentially level with GPT-5.4 at 67.4 and ahead of Claude Opus 4.6 at 60.1. On the same overall measure, previous-generation Qwen3.6-27B scored 42.2.
Continue reading
The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in✓ Signed in — this article isn’t included in your current plan.
Unlocking the full article...