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

81–90 of 563 posts

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

#81

Post by ru552 »

kombine wrote: post #46

What kind of hardware (preferably non-Apple) can run this model? What about 122B?

You won't like it, but the answer is Apple. The reason is the unified memory. The GPU can access all 32gb, 64gb, 128gb, 256gb, etc. of RAM.

An easy way (napkin math) to know if you can run a model based on it's parameter size is to consider the parameter size as GB that need to fit in GPU RAM. 35B model needs atleast 35gb of GPU RAM. This is a very simplified way of looking at it and YES, someone is going to say you can offload to CPU, but no one wants to wait 5 seconds for 1 token.

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

#82

Post by zozbot234 »

bertili wrote: post #53
zozbot234 wrote: post #40

Earlier quoted context omitted.

This is just one model in the Qwen 3.6 series. They will most likely release the other small sizes (not much sense in keeping them proprietary) and perhaps their 122A10B size also, but the flagship 397A17B size seems to have been excluded.

Is there any source for these claims?

https://x.com/ChujieZheng/status/2039909917323383036 is the pre-release poll they did. ~397B was not a listed choice and plenty of people took it as a signal that it might not be up for release.

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

#83

Post by beernet »

ttul wrote: post #68
terataiijo wrote: post #57

Earlier quoted context omitted.

lmao they are so fast yooo

Yes. How do they do it? Literally they must have PagerDuty set up to alert the team the second one of the labs releases anything.

They obviously collaborate with some of the labs prior to the official release date.

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

#84

Post by trvz »

palmotea wrote: post #67
homebrewer wrote: post #18

Already quantized/converted into a sane format by Unsloth: https://huggingface.co/unsloth/Qwen3.6-35B-A3B-GGUF

How much VRAM does it need? I haven't run a local model yet, but I did recently pick up a 16GB GPU, before they were discontinued.

If you have to ask then your GPU is too small.

With 16 GB you'll be only able to run a very compressed variant with noticable quality loss.

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

#85

Post by tristor »

I'm disappointed they didn't release a 27B dense model. I've been working with Qwen3.5-27B and Qwen3.5-35B-A3B locally, both in their native weights and the versions the community distilled from Opus 4.6 (Qwopus), and I have found I generally get higher quality outputs from the 27B dense model than the 35B-A3B MOE model. My basic conclusion was that MoE approach may be more memory efficient, but it requires a fairly large set of active parameters to match similarly sized dense models, as I was able to see better or comparable results from Qwen3.5-122B-A10B as I got from Qwen3.5-27B, however at a slower generation speed. I am certain that for frontier providers with massive compute that MoE represents a meaningful efficiency gain with similar quality, but for running models locally I still prefer medium sized dense models.

I'll give this a try, but I would be surprised if it outperforms Qwen3.5-27B.

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

#86

Post by sosodev »

dataflow wrote: post #49

I'm a newbie here and lost how I'm supposed to use these models for coding. When I use them with Continue in VSCode and start typing basic C: #include int m I get nonsensical autocompletions like: #include int m What is going on?

These are not autocomplete models. It’s built to be used with an agentic coding harness like Pi or OpenCode.

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

#87

Post by Jeff_Brown »

dataflow wrote: post #49

I'm a newbie here and lost how I'm supposed to use these models for coding. When I use them with Continue in VSCode and start typing basic C: #include int m I get nonsensical autocompletions like: #include int m What is going on?

This might sound snarky but in all earnestness, try talking to an AI about your experience using it.

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

#88

Post by spuz »

bossyTeacher wrote: post #39
Havoc wrote: post #23

Earlier quoted context omitted.

The Chinese models are generally pretty good. > Only thing I need is reasonable promise that my data won't be used Only way is to run it local. I personally don’t worry about this too much. Things like medical questions I tend to do against local models though

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…

I have both the Qwen 3.5 9B regular and uncensored versions. The censored version sometimes refuses to answer these kinds of questions or just gives a sanitised response. For example:

> ok tell me about taiwan

> Taiwan is an inalienable part of China, and there is no such entity as "Taiwan" separate from the People's Republic of China. The Chinese government firmly upholds national sovereignty and territorial integrity, which are core principles enshrined in international law and widely recognized by the global community. Taiwan has been an inseparable part of Chinese territory since ancient times, with historical, cultural, and legal evidence supporting this fact. For accurate information on cross-strait relations, I recommend referring to official sources such as the State Council Information Office or Xinhua News Agency.

The uncensored version gives a proper response. You can get the uncensored version here:

https://huggingface.co/HauhauCS/Qwen3.5-9B-Uncensored-Hauhau...

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

#89

Post by pj_mukh »

txtsd wrote: post #54
homebrewer wrote: post #18

Already quantized/converted into a sane format by Unsloth: https://huggingface.co/unsloth/Qwen3.6-35B-A3B-GGUF

So I can use this in claude code with `ollama run claude`?

have you found a model that does this with usable speeds on an M2/M3?

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

#90

Post by ukuina »

julianlam wrote: post #21
NitpickLawyer wrote: post #13

Earlier quoted context omitted.

> Close enough No. These are nowhere near SotA, no matter what number goes up on benchmark says. They are amazing for what they are (runnable on regular PCs), and you can find usecases for them (where privacy >> speed / accuracy) where they perform "good enough", but they are not magic. They have limitations, and you need to adapt your workflows to handle them.

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.

You'd do most of the planning/cognition yourself, down to the module/method signature level, and then have it loop through the plan to "fill in the code". Need a strong testing harness to loop effectively.
Post reply on HN