News for hackers. Stuff that matters.

A comment on Qwen 3.8 27B on my laptop can interpret model weights, too. Parent comment

Re:Qwen 3.8 27B on my laptop can interpret model weights, too by t0mas88 on Tuesday November 03, 2026 @05:00PM (#50358829)
Realistically a 32GB card is what you need to run Gemma4:31b well. Doing it in 24GB leaves almost no room for context / KV cache.
[ Parent | on HN ]
Re:Qwen 3.8 27B on my laptop can interpret model weights, too by ActorNightly on Tuesday November 03, 2026 @10:13PM (#50361120)
32k context is plenty fine. thats 50 pages of single spaced text
[ Parent | on HN ]
Re:Qwen 3.8 27B on my laptop can interpret model weights, too by leansensei on Friday November 06, 2026 @02:45PM (#50397388)
Not if you preserve reasoning in the KV cache, 32k ctx is nothing.
[ Parent | on HN ]