Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Qwen 3.6 27B is the sweet spot for local development

quesma.com

Post reply on HN
Print view View original

531–540 of 809 posts

Re: Qwen 3.6 27B is the sweet spot for local development

#531

Post by indemnity »

FuckButtons wrote: post #528
JeremyNT wrote: post #121

Earlier quoted context omitted.

I also don't understand why people in this price bracket are buying Mac laptops instead of desktop computers with GPUs? Just to flex that it's portable?

The fact that I can take it with me? That I don’t need internet to still have access to deepseek? The fact that electricity is expensive and an mbp uses ~10% of the power that an equivalent vram set up would using gpu’s. Also, in order to get the same vram I would need to spend a similar amount, but wouldn’t also have a machine that was useful for other workloads that need a huge amount of ram.

Potentially going to sound privileged here, but why not both?

Personally when going on the road I like portability (14" MBP or MBA), but at home I want raw non-thermally throttled power.

Re: Qwen 3.6 27B is the sweet spot for local development

#532

Post by throwaway240403 »

iagooar wrote: post #199

I love my MacBook Pro M5 128GB RAM and I love qwen3.6. BUT DO NOT buy this MacBook if you plan on doing serious coding using local LLMs with it. The reason is simple: your fingers will burn and your head will explode from the noise. Running any kind of sophisticated job on the very laptop you are using is just not viable. Sure you can use it in clamshell mode, but forget touching it while working with AI coding or ag…

No, buy a framework desktop.

Re: Qwen 3.6 27B is the sweet spot for local development

#533

Post by satvikpendem »

JeremyNT wrote: post #121
doodlesdev wrote: post #83

I feel like I'm going insane seeing people buy these 128gb MBP for thousands of dollars to run models that are objectively much worse than SOTA and spending so much more. The amount spent on a 128gb M5 MAX can buy you a damned new car here. What the hell am I missing? Are developers in other countries living in such different worlds? (I'm aware the price is, in absolute terms, more expensive where I live compared to…

I also don't understand why people in this price bracket are buying Mac laptops instead of desktop computers with GPUs? Just to flex that it's portable?

Unified memory.

Re: Qwen 3.6 27B is the sweet spot for local development

#534

Post by dgacmu »

827a wrote: post #518
iagooar wrote: post #199

I love my MacBook Pro M5 128GB RAM and I love qwen3.6. BUT DO NOT buy this MacBook if you plan on doing serious coding using local LLMs with it. The reason is simple: your fingers will burn and your head will explode from the noise. Running any kind of sophisticated job on the very laptop you are using is just not viable. Sure you can use it in clamshell mode, but forget touching it while working with AI coding or ag…

Apple does not sell a 64GB variant of the M4 Mac Mini. IIRC they never have; its always capped out at 48GB. If you were planning on getting an M5 128GB; just get a DGX Spark (~$4500) or a 5090-equipped machine (~$4500) plus a Macbook Air (~$1500). You'll come in below the M5 Max 128 pricing (~$6700+ USD) and be happier for it.

That's incorrect, I have one on my desk right now. They've stopped selling it now, but I got one a year and a half ago:

> Apple M4 Pro chip with 14‑core CPU, 20‑core GPU, 16-core Neural Engine 64GB unified memory 2TB SSD storage 10 Gigabit Ethernet Three Thunderbolt 5 ports, HDMI port, two USB‑C ports, headphone jack Accessory Kit $2,649.00

Re: Qwen 3.6 27B is the sweet spot for local development

#536

Post by trey-jones »

Qwen3.6 was the first model I ran locally that seemed smart, but qwen3-coder:30b is way, way more responsive and more accurate for writing code according to my tests, including human-eval. If you can run one than you can almost certainly run the other. If you haven't tried qwen3-coder I would definitely recommend it. It feels positively snappy compared to every other local model I've tried. All you need is 32G VRAM and some heat dissipation.

Re: Qwen 3.6 27B is the sweet spot for local development

#537

Post by janalsncm »

hollowturtle wrote: post #387
janalsncm wrote: post #100

Earlier quoted context omitted.

> Being able to nail a zero-shot greenfield project is relatively easy even for a small model Not really germane to your comment but I hope I don’t sound old when I say I remember a time when spinning up a PoC was a week of work, and a statement like yours was pure science fiction.

In what era spinning up a PoC required a week of work? Especially on the web. I've been a developer for roughly 20 years and that has never been the case, to the point that I believe people impressed by LLMs are the same who had a very low productivity. Today we have game jams as short as 3 days and talented people are able to produce very good PoC, with some almost complete!

1) It depends entirely on the concept you are trying to prove and how experienced you are in that domain.

2) Not every team will have someone with 20 years of experience in a particular domain eager to spin up a PoC.

Re: Qwen 3.6 27B is the sweet spot for local development

#538

Post by zargon »

naikrovek wrote: post #463
petilon wrote: post #36

Earlier quoted context omitted.

Memory requirements of newer models will increase, so while the hardware may last 10 years it won't be able to run the latest models for 10 years.

Available models aren’t really trending upward in size. Not like I thought they would, anyway. They’re trending to be the right size to be good. Qwen3.6-35B is not as good as Qwen3.6-27B. The larger model is faster, but a lot dumber; it gets caught in loops, makes crazy mistakes, and is just not as good. It’s bigger, but it is nowhere near as good as the 27B variant.

Qwen3.6-35B-A3B is worse than 27B because it's an MoE and 27B is dense. 35B only passes each token through 3B of its total parameters, whereas 27B sends each token through all 27B parameters.

Re: Qwen 3.6 27B is the sweet spot for local development

#539

Post by 2Gkashmiri »

iagooar wrote: post #199

I love my MacBook Pro M5 128GB RAM and I love qwen3.6. BUT DO NOT buy this MacBook if you plan on doing serious coding using local LLMs with it. The reason is simple: your fingers will burn and your head will explode from the noise. Running any kind of sophisticated job on the very laptop you are using is just not viable. Sure you can use it in clamshell mode, but forget touching it while working with AI coding or ag…

How is Mac studio 32gb or 96 gb ram one?

Re: Qwen 3.6 27B is the sweet spot for local development

#540

Post by iosjunkie »

neonstatic wrote: post #330
acchow wrote: post #202

Earlier quoted context omitted.

That $6700 is a $5000 upgrade over a base model Macbook Pro. $5000 in US Treasuries (currently at 4.89%) yields $244.5/yr. That's more than enough to cover the annual Claude Pro subscription ($200/yr) which includes Claude Code with lots of Sonnet usage (far better than Qwen 3.6)

I think the argument isn't that local is cheaper - it's that local is doable and delivers unparalleled privacy.

And your government can’t take it away on a Friday afternoon.
Post reply on HN