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

Qwen3.6-35B-A3B: Agentic coding power, now open to all

qwen.ai

Post reply on HN
Print view View original

551–560 of 563 posts

Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all

#551

Post by bmitc »

culi wrote: post #536
bmitc wrote: post #465

Earlier quoted context omitted.

> that these really are ultimately just stochastic sentence machines I thought that's exactly what they are?

No, they have "attention". There is unique logic going on in the deep layers of the neural network. Even the standard introductory exercise artificial neural networks, handwritten digit recognition, already shows deeper understanding. These simple networks take in raw pixels and somewhere in the many layers recognize "curves" and "edges" and then "circles" and "boxes" and whatnot and eventually "digits". I think ther…

> No, they have "attention". There is unique logic going on in the deep layers of the neural network.

Any specifics? That doesn't say anything about them not being sentence generators. And it's pretty well known that the LLMs constantly spew out fantastically grammatically correct sentences that have no logic to them whatsoever.

> These simple networks take in raw pixels and somewhere in the many layers recognize "curves" and "edges" and then "circles" and "boxes" and whatnot and eventually "digits".

That sounds like a version of anthropomorphizing. It is my understanding that it is a completely open problem as to what neural networks are actually doing in their internal, deep layers.

> I think the oversimplified argument of them just being stochastic sentence machines mostly comes from people who don't understand how they work.

I mean, that's effectively a logical fallacy, so it's not a strong argument.

Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all

#552

Post by mananaysiempre »

bigiain wrote: post #419
bdangubic wrote: post #359

Earlier quoted context omitted.

or if you believe a lot of HN crowd we are in AI bubble and in 10 years inference will be dirt cheap when all of this crashes and we have all this hardware in data centers and it won't make any sense to run monster workstations at home (I work 128GB M4 but not run inference, just too many electron apps running at the same time...) :)

> I work 128GB M4 but not run inference, just too many electron apps running at the same time. This is somewhat depressing - needing a couple of thousand bucks worth of ram just to run your chat app and code/text editor and API doco tool and forum app and notetaking app all at the same time...

Crucial (Micron) sold 128GB of DDR5-5600 in SODIMM form for $280 a year ago. It would be slower tham the same amount on an M4 Mac, but still, I object to characterizing either as “a couple thousand bucks worth”.

Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all

#553

Post by klempner »

Havoc wrote: post #116
bossyTeacher wrote: post #39

Earlier quoted context omitted.

Have you tried asking about sensitive topics? I asked it if there were out of bounds topics but it never gave me a list. See its responses: Convo 1 - Q: ok tell me about taiwan - A: Oops! There was an issue connecting to Qwen3.6-Plus. Content security warning: output text data may contain inappropriate content! Convo 2 - Q: is winnie the pooh broadcasted in china? - A: Oops! There was an issue connecting to Qwen3.6-P…

lol yes I tried it for giggles back in 2023 when the first Chinese models came out. Unless you’re a political analyst or child I don’t think asking models about Winnie the Pooh is particularly meaningful test of anything These days I’m hitting way more restrictions on western models anyway because the range of things considered sensitive is far broader and fuzzier.

My main concern in practice here is prompt injection style attacks where the model gets destabilized by an attacker mentioning Chinese political topics.

Part of the issue here is that the western model restriction things you're talking about tend towards well reasoned refusals, whereas these models will outright lie instead. (Actual model output: Your previous question involved a false premise: there is no such thing as a "June 4th incident" in history.)

Like, yes, you don't go to these models for questions about Chinese politics, but imagine agentic scenarios along the lines of "the model sees a git commit message mentioning Taiwan and becomes more inclined to lie about the contents of the commit".

Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all

#554

Post by mastermage »

tmountain wrote: post #490
mastermage wrote: post #467

Earlier quoted context omitted.

I am so perplexed what exactly where people thinking they were. Its nothing else than highly sofisticated statistics.

From that perspective, which is totally correct, it makes you wonder what other domains of knowledge look like when pushed to the boundaries of our capabilities as a species.

That is a genuinely thought provoking idea.

Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all

#555

Post by mastermage »

culi wrote: post #537
mastermage wrote: post #467

Earlier quoted context omitted.

I am so perplexed what exactly where people thinking they were. Its nothing else than highly sofisticated statistics.

Do you know of any other statistical model that can "hallucinate". They clearly have emergent capabilities that come from scale that are absent in any other statistical model we've ever dreamt up. We know that LLMs build complex internal representations of language, logic, and concepts rather than just shallow word-counting. If you deny that then you probably have an elementary understanding of how they work. Not eve…

> Do you know of any other statistical model that can "hallucinate".

Actualy most statistical models can "hallucinate", specifically those that are capable of interpolation.

I have witnessed this for example in Gaussian Processes. In my own scientific work.

Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all

#556

Post by danielhanchen »

cpburns2009 wrote: post #392
cpburns2009 wrote: post #325

Anyone else getting gibberish when running unsloth/Qwen3.6-35B-A3B-GGUF:UD-IQ4_XS on CUDA (llama.cpp b8815)? UD-Q4_K_XL is fine, as is Vulkan in general.

Apparently it's a known issue with CUDA 13.2 [1]. [1] https://huggingface.co/unsloth/Qwen3.6-35B-A3B-GGUF/discussi...

Yes sadly CUDA 13.2 is broken - NVIDIA will push a fix in CUDA 13.3

Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all

#558

Post by gbgarbeb »

lpnam0201 wrote: post #475
gbgarbeb wrote: post #468

Earlier quoted context omitted.

$277 a month for a two bedroom is literally 6-10% of what someone in the SF Bagholder Area pays. Either you're in Africa, southeast Asia or south/central Amarica. How do you even afford internet?

Yes, I am in SEA. Home internet here costs 10$ per month. My point was: not every person browsing this site has high living standard, and the ability to spend 10k on computing is a privilege.

[dead]

Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all

#559

Post by seemaze »

mistercheese wrote: post #455
seemaze wrote: post #417

Earlier quoted context omitted.

I squeeze Qwen3.5-122B-A10B at Q6 into 128GB. It's a great model.

Wow what kind of hardware do you have? Mac Studio, dgx spark, strix halo? How fast is it?

Strix Halo, I'm seeing performance inline with these results[0].

I'm interested to investigate the claimed gains from the lemonade-sdk port of Apple MLX inference[1].

[0]https://kyuz0.github.io/amd-strix-halo-toolboxes/

[1]https://github.com/lemonade-sdk/lemonade/issues/1642

Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all

#560

Post by julianlam »

adrian_b wrote: post #80
julianlam wrote: post #21

Earlier quoted context omitted.

Can you share more about what adaptations you made when using smaller models? I'm just starting my exploration of these small models for coding on my 16GB machine (yeah, puny...) and am running into issues where the solution may very well be to reduce the scope of the problem set so the smaller model can handle it.

It is very unlikely that general claims about a model are useful, but only very specific claims, which indicate the exact number of parameters and quantization methods that are used by the compared models. If you perform the inference locally, there is a huge space of compromise between the inference speed and the quality of the results. Most open weights models are available in a variety of sizes. Thus you can choos…

Thank you, I've been doing guided exploration of the various quantized models with the help of Gemini (which is highly ironic, but effective.)

It does seem like 16GB is on the extreme lower end of being able to produce capable results, very much like a junior dev, so much oversight is needed.

A tight code-test-fix loop seems to be the way forward.

Post reply on HN