Qwen3.6-35B-A3B: Agentic Coding Power, Now Open to All

Re:Qwen3.6-35B-A3B: Agentic Coding Power, Now Open to All by bwv848 on Thursday April 16, 2026 @05:00PM (#47799450)

I've been trying the Q4_K_M version, and sometimes it gets stuck in a loop. Gemma 4 doesn’t have this issue.

[ Parent | on HN ]

Re:Qwen3.6-35B-A3B: Agentic Coding Power, Now Open to All by yencabulator on Thursday April 16, 2026 @05:33PM (#47799772)

This has happened before with quantizations and other backends (ones not used by the research lab). Give it a week, download latest versions of everything, and try again.

[ Parent | on HN ]

Re:Qwen3.6-35B-A3B: Agentic Coding Power, Now Open to All by Readerium on Thursday April 16, 2026 @09:10PM (#47801493)

perhaps increasing repitition_penalty might be helpful

[ Parent | on HN ]

Re:Qwen3.6-35B-A3B: Agentic Coding Power, Now Open to All by mobiuscog on Friday April 17, 2026 @08:22AM (#47805142)

I'm having the same issues, the more I use it. The repetition penalty doesn't seem to help.

I get some really amusing 'reflective' responses, but I think it needs a bit more cooking. Maybe I'll try another variant.

[ Parent | on HN ]