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

Qwen3.6-27B: Flagship-Level Coding in a 27B Dense Model

qwen.ai

Post reply on HN
Print view View original

381–390 of 482 posts

Re: Qwen3.6-27B: Flagship-Level Coding in a 27B Dense Model

#381

Post by solenoid0937 »

swiftcoder wrote: post #249
fnordpiglet wrote: post #84

Earlier quoted context omitted.

For coding often quality at the margin is crucial even at a premium. It’s not the same as cranking out spam emails or HN posts at scale. This is why the marginal difference between your median engineer and your P99 engineer is comp is substantial, while the marginal comp difference between your median pick and packer vs your P99 pick and packer isn’t. I’d also say it keeps the frontier shops competitive while costing…

> For coding often quality at the margin is crucial even at a premium For some problems, sure, and when you are stuck, throwing tokens at Opus is worthwhile. On the other hand, a $10/month minimax 2.7 coding subscription that literally never runs out of tokens will happily perform most day-to-day coding tasks

"Literally never runs out of tokens?" lol, no. Tokens are just energy. There is always a way to run out of tokens, and no one will subsidize free tokens forever.

Re: Qwen3.6-27B: Flagship-Level Coding in a 27B Dense Model

#382

Post by otabdeveloper4 »

fnordpiglet wrote: post #197
otabdeveloper4 wrote: post #125

Earlier quoted context omitted.

> For coding often quality at the margin is crucial even at a premium. For coding, quality is not measurable and is based entirely on feels (er, sorry, "vibes"). Employers paying for SOTA models is nothing but a lifestyle status perk for employees, like ping-pong tables or fancy lunch snacks.

I’m building my own company and I consider model choice crucial to my marginal ability to produce a higher quality product I don’t regret having built. Every higher end dev shop I’ve worked at over the last few years perceives things the same. There are measurable outcomes from software built well and software not, even if the code itself isn’t easily measurable. I would rather pay a few thousand more per year for a…

"Rational engineering and business choice" and "AI" are two words that do not go together.

Wait five years and come back. Right now AI is 100% FOMO and lifestyle signaling and nothing more.

Re: Qwen3.6-27B: Flagship-Level Coding in a 27B Dense Model

#383

Post by czl »

proxysna wrote: post #284
tandr wrote: post #79

Earlier quoted context omitted.

What would be these additional vllm flags, if you don't mind sharing?

This is from an example from my Nomad cluster with two a5000's, which is a bit different what i have at work, but it will mostly apply to most modern 24G vram nvidia gpu. "--tensor-parallel-size", "2" - spread the LLM weights over 2 GPU's available "--max-model-len", "90000" - I've capped context window from ~256k to 90k. It allows us to have more concurrency and for our use cases it is enough. "--kv-cache-dtype", "f…

> "--speculative-config",

Regarding that last option: speculation helps max concurrency when it replaces many memory-expensive serial decode rounds with fewer verifier rounds, and the proposer is cheap enough. It hurts when you are already compute-saturated or the acceptance rate is too low. Good idea to benchmark a workload with and without speculative decoding.

Re: Qwen3.6-27B: Flagship-Level Coding in a 27B Dense Model

#384

Post by rohansood15 »

fnordpiglet wrote: post #149
rohansood15 wrote: post #91

Earlier quoted context omitted.

Most code is not P99 though. Also, have you considered that your trust in Anthropic and distrust in China may not be shared by many outside the US? There's a reason why Huawei is the largest supplier of 5G hardware globally.

I find it hard to believe anyone who has ever done business inside China doesn’t know that the structure of Chinese business is built around massive IP theft and repurposing on a state wide systematic level. It’s not a nationalism point, it’s an objective and easily verified truth. Most code is not P99, but companies pay a premium to produce code that is. That’s my point.

Chinese companies are built on IP theft, and Anthropic/Open AI are not?

And in world where code generation costs are trending to zero, goodluck commanding a premium to produce any kind of code.

There is a whole bunch of P99 code that is open-source. What makes code P99 is not the model that produces it, but the people who verify/validate/direct it.

Re: Qwen3.6-27B: Flagship-Level Coding in a 27B Dense Model

#385

Post by petu »

mft_ wrote: post #271

Huh, running the Q4_K_M quant with LM Studio, and asked it "How can I set up Qwen 3.6 27b to use tools and access the local file system?". Part of its reply was: Quick clarification: As of early 2025, "Qwen 3.6" hasn't been released yet. You are likely looking for Qwen2.5, specifically the Qwen2.5-32B-Instruct model, which is the 30B-class model closest to your 27B reference. The instructions below will use this mode…

Models are math functions that predict next word, not conscious beings. If it was trained on dataset including data up to Q1 2025, then that's more or less expected answer -- even Qwen 3 didn't exist.

If you see model that can reliably answer questions about itself (version, family, capabilities, etc), then it's most likely part of system prompt.

In absence of system prompt even Claude could say it's a model created by DeepSeek: https://x.com/stevibe/status/2026227392076018101

Re: Qwen3.6-27B: Flagship-Level Coding in a 27B Dense Model

#386

Post by zozbot234 »

adrian_b wrote: post #330
anonym29 wrote: post #316

Earlier quoted context omitted.

That sounds like a huge hassle for what I imagine must be peak speeds of low double digit tok/s PP and TG, even with effective prompt caching and self-ngram and all the other tricks, no? If I really feel like I needed larger models locally (I don't, the 120/122B A10/12B models are awesome on my hardware), I think I'd rather just either pony up for a used M3 Ultra 512GB, wait for an M5 Ultra (hoping they bring back 51…

Until now I have not run models that do not fit in 128 GB. I have an Epyc server with 128 GB of high-throughput DRAM, which also has 2 AMD GPUs with 16 GB of DRAM each. Until now I have experimented only with models that can fit in this memory, e.g. various medium-size Qwen and Gemma models, or gpt-oss. But I am curious about how bigger models behave, e.g. GLM-5.1, Qwen3.5-397B-A17B, Kimi-K2.6, DeepSeek-V3.2, MiniMax…

AIUI, the main obstacle to maximizing performance with SSD offload is that existing GGUF files for MoE models are not necessarily laid out so that fetching a single MoE layer-expert can be done by reading a single sequential extent off the file. It may be that the GGUF format is already flexible enough in its layout configuration that this is doable with a simple conversion; but if not, the GGUF specification would have to be extended to allow such a layout to be configured.

Re: Qwen3.6-27B: Flagship-Level Coding in a 27B Dense Model

#387

Post by Alifatisk »

simonw wrote: post #106

The pelican is excellent for a 16.8GB quantized local model: https://simonwillison.net/2026/Apr/22/qwen36-27b/ I ran it on an M5 Pro with 128GB of RAM, but it only needs ~20GB of that. I expect it will run OK on a 32GB machine. Performance numbers: Reading: 20 tokens, 0.4s, 54.32 tokens/s Generation: 4,444 tokens, 2min 53s, 25.57 tokens/s I like it better than the pelican I got from Opus 4.7 the other day: https://si…

So this is it. We have finally achieved excellent illustrating of your svg art.

Re: Qwen3.6-27B: Flagship-Level Coding in a 27B Dense Model

#388

Post by Mashimo »

refulgentis wrote: post #141
Marciplan wrote: post #93

Earlier quoted context omitted.

every model release Simon comes with his Pelican and then this comment follows. Can we stop both? its so boring

I really appreciate you speaking up. Happened yesterday on GPT Image 2, bit my tongue b/c people would see it as fun policing, and same thing today. And it happens on every. single. LLM. release. thread. It's disruptive to the commons, doesn't add anything to knowledge of a model at this point, and it's way out of hand when people are not only engaging with the original and creating screenfuls to wade through before…

> and creating screenfuls to wade through before on-topic content,

It's often just a single root comment that you can collapse.

I find how svg drawing skills improve over time interesting. Very simple and very small datapoint. But I still find value in it.

Re: Qwen3.6-27B: Flagship-Level Coding in a 27B Dense Model

#389

Post by jedisct1 »

datadrivenangel wrote: post #279

So far I'm unimpressed for local inference. got 11 tokens per second on omlx on an M5 Pro with 128gb of ram, so it took an hour to write a few hundred lines of code that didn't work. Opus and Sonnet in CC the same task successfully in a matter of minutes. The 3.6:35b model seemed okay on ollama yesterday. Need to check out other harnesses for this besides claude code, but the local models are just painfully slow.

For local models, you should check out https://swival.dev instead of Claude Code.

Re: Qwen3.6-27B: Flagship-Level Coding in a 27B Dense Model

#390

Post by hnfong »

bigyabai wrote: post #336
reddit_clone wrote: post #333

What can I run on a M4 Pro with 48 GB or RAM?

A sparser model like Qwen3.6 35B A3B is probably your best choice: https://qwen.ai/blog?id=qwen3.6-35b-a3b

The 35B MOE will run faster, but 48GB RAM is more than enough to run the 27B dense model as well. It's just that token/s will be on the lower side.
Post reply on HN