Qwen 3.6 27B is the sweet spot for local development

Re:Qwen 3.6 27B is the sweet spot for local development by kristianp on Monday June 29, 2026 @04:39PM (#48724895)

You're only going to get an incremental improvement with an M5 Pro mini compared to an M4 Pro mini. Memory bandwidth goes from 273GB/s to 307GB/s, about 12.5% improvement for LLMs.

[ Parent | on HN ]

Re:Qwen 3.6 27B is the sweet spot for local development by freehorse on Monday June 29, 2026 @06:19PM (#48726080)

M5's have the neural accelarator that boosts prefill speed a lot. But token generation itself will not change that much, that's true.

[ Parent | on HN ]

Re:Qwen 3.6 27B is the sweet spot for local development by iagooar on Monday June 29, 2026 @04:41PM (#48724920)

I thought they might ship an M5 Max version, but you are probably right.

[ Parent | on HN ]