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

Re:Qwen3.6-35B-A3B: Agentic Coding Power, Now Open to All by Aurornis on Thursday April 16, 2026 @11:28AM (#47794681)

Additional VRAM is needed for context.

This model is a MoE model with only 3B active parameters per expert which works well with partial CPU offload. So in practice you can run the -A(N)B models on systems that have a little less VRAM than you need. The more you offload to the CPU the slower it becomes though.

[ Parent | on HN ]

Re:Qwen3.6-35B-A3B: Agentic Coding Power, Now Open to All by Glemllksdf on Thursday April 16, 2026 @11:52AM (#47795187)

Isn't that some kind of gambling if you offload random experts onto the CPU?

Or is it only layers but that would affect all Experts?

[ Parent | on HN ]

Re:Qwen3.6-35B-A3B: Agentic Coding Power, Now Open to All by dragonwriter on Thursday April 16, 2026 @12:22PM (#47795724)

Pretty sure all partial offload systems I’ve seen work by layers, but there might be something else out there.

[ Parent | on HN ]

Re:Qwen3.6-35B-A3B: Agentic Coding Power, Now Open to All by namibj on Friday April 17, 2026 @10:55AM (#47806626)

Speculative decoding is already gambling.

[ Parent | on HN ]