Qwen 3.8 27B on my laptop can interpret model weights, too

Re:Qwen 3.8 27B on my laptop can interpret model weights, too by justinclift on Tuesday November 03, 2026 @03:27PM (#50357718)

> Qwen3.8-Flash_Next should be in the running for that title as well.

Unfortunately it still hallucinates, which limits it's usefulness for real world tasks. :(

[ Parent | on HN ]

Re:Qwen 3.8 27B on my laptop can interpret model weights, too by bityard on Tuesday November 03, 2026 @03:50PM (#50357988)

Every model hallucinates, if not given the right context and tools.

[ Parent | on HN ]

Re:Qwen 3.8 27B on my laptop can interpret model weights, too by komali2 on Tuesday November 03, 2026 @09:58PM (#50361036)

How do you tailor your context and what tools do you use?

[ Parent | on HN ]

Re:Qwen 3.8 27B on my laptop can interpret model weights, too by hasperdi on Wednesday November 04, 2026 @05:25AM (#50363572)

give it information it needs to complete it's task, and not more so it doesn't get confused.

[ Parent | on HN ]

Re:Qwen 3.8 27B on my laptop can interpret model weights, too by justinclift on Wednesday November 04, 2026 @05:50AM (#50363763)

I find they hallucinate even with the right context and tools.

So, while useful for some people, for the things I care about they're presently pretty useless. :(

[ Parent | on HN ]