any tips for running it locally within an agent harness? maybe using pi or opencode?
Qwen3.6-35B-A3B: Agentic coding power, now open to all
351–360 of 563 posts
Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all
#352Earlier quoted context omitted.
$10k is well outside my budget for frivolous computer purchases.
99.97% of HN users are nodding… :)
There are way too many good uses of these models for local that I fully expect a standard workstation 10 years from now to start at 128GB of RAM and have at least a workstation inference device.
Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all
#353Earlier 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.
How many people/hackernews can run a 397b param model at home? Probably like 20-30.
Running the mxfp4 unsloth quant of qwen3.5-397b-a17b, I get 40 tps prefill, 20tps decode.
AMD threadripper pro 9965WX, 256gb ddr5 5600, rtx 4090.
Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all
#354I've been running this on my laptop with the Unsloth 20.9GB GGUF in LM Studio: https://huggingface.co/unsloth/Qwen3.6-35B-A3B-GGUF/blob/mai... It drew a better pelican riding a bicycle than Opus 4.7 did! https://simonwillison.net/2026/Apr/16/qwen-beats-opus/
I wonder when pelican riding a bicycle will be useless as an evaluation task. The point was that it was something weird nobody had ever really thought about before, not in the benchmarks or even something a team would run internally. But now I'd bet internally this is one of the new Shirley Cards.
Yeah try it with something else, or e.g. add a tiger to the back seat.
Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all
#355Earlier quoted context omitted.
I mean look at the result where he asked about a unicycle - the model couldn't even keep the spokes inside the wheels - would be rudimentary if it "learned" what it means to draw a bicycle wheel and could transfer that to unicycle.
it's the frame that's surprisingly - and consistentnly - wrong. You'd think two triangles would be pretty easy to repro; once you get that the rest is easy. It's not like he's asking "draw a pelican on a four-bar linkage suspension mountainbike..."
This is older, but even humans don't have a great concept of how a bicycle works... https://twistedsifter.com/2016/04/artist-asks-people-to-draw...
Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all
#356I have been using Qwen3.5-35B-A3B a lot in local testing, and it is by far the most capable model that could fit on my machine.
I think quantization technology has really upped its game around these models,
and there were two quants that blew me away
Mudler APEX-I-Quality. then later I tried Byteshape Q3_K_S-3.40bpw
Both made claims that seemed too good to be true, but I couldn't find any traces of lobotomization doing long agent coding loops. with the byteshape quant I am up to 40+ t/s which is a speed that makes agents much more pleasant. On an rtx 3060 12GB and 32GB of system ram, I went from slamming all my available memory to having like 14GB to spare.
Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all
#357Earlier quoted context omitted.
It's on the page: Precision Quantization Tag File Size 1-bit UD-IQ1_M 10 GB 2-bit UD-IQ2_XXS 10.8 GB UD-Q2_K_XL 12.3 GB 3-bit UD-IQ3_XXS 13.2 GB UD-Q3_K_XL 16.8 GB 4-bit UD-IQ4_XS 17.7 GB UD-Q4_K_XL 22.4 GB 5-bit UD-Q5_K_XL 26.6 GB 16-bit BF16 69.4 GB
I really want to know what does M, K, XL XS mean in this context and how to choose. I searched all unsloth doc and there seems no explaination at all.
Q4_K is a type of quantization. It means that all weights will be at a minimum 4bits using the K method.
But if you're willing to give more bits to only certain important weights, you get to preserve a lot more quality for not that much more space.
The S/M/L/XL is what tells you how many tensors get to use more bits.
The difference between S and M is generally noticeable (on benchmarks). The difference between M and L/XL is less so, let alone in real use (ymmv).
Here's an example of the contents of a Q4_K_:
S
llama_model_loader: - type f32: 392 tensors
llama_model_loader: - type q4_K: 136 tensors
llama_model_loader: - type q5_0: 43 tensors
llama_model_loader: - type q5_1: 17 tensors
llama_model_loader: - type q6_K: 15 tensors
llama_model_loader: - type q8_0: 55 tensors
M
llama_model_loader: - type f32: 392 tensors
llama_model_loader: - type q4_K: 106 tensors
llama_model_loader: - type q5_0: 32 tensors
llama_model_loader: - type q5_K: 30 tensors
llama_model_loader: - type q6_K: 15 tensors
llama_model_loader: - type q8_0: 83 tensors
L
llama_model_loader: - type f32: 392 tensors
llama_model_loader: - type q4_K: 106 tensors
llama_model_loader: - type q5_0: 32 tensors
llama_model_loader: - type q5_K: 30 tensors
llama_model_loader: - type q6_K: 14 tensors
llama_model_loader: - type q8_0: 84 tensorsRe: Qwen3.6-35B-A3B: Agentic coding power, now open to all
#358I've been running this on my laptop with the Unsloth 20.9GB GGUF in LM Studio: https://huggingface.co/unsloth/Qwen3.6-35B-A3B-GGUF/blob/mai... It drew a better pelican riding a bicycle than Opus 4.7 did! https://simonwillison.net/2026/Apr/16/qwen-beats-opus/
I'm not sure how you can give the flamingo win to Qwen:
* It's sitting on the tire, not the seat.
* Is that weird white and black thing supposed to be a beak? If so, it's sticking out of the side of its face rather than the center.
* The wheel spokes are bizarre.
* One of the flamingo's legs doesn't extend to the pedal.
* If you look closely at the sunglasses, they're semi-transparent, and the flamingo only has one eye! Or the other eye is just on a different part of its face, which means the sunglasses aren't positioned correctly. Or the other eye isn't.
* (subjective) The sunglasses and bowtie are cute, but you didn't ask for them, so I'd actually dock points for that.
* (subjective) I guess flamingos have multiple tail feathers, but it looks kinda odd as drawn.
In contrast, Opus's flamingo isn't as detailed or fancy, but more or less all of it looks correct.
Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all
#359Earlier quoted context omitted.
99.97% of HN users are nodding… :)
There are way too many good uses of these models for local that I fully expect a standard workstation 10 years from now to start at 128GB of RAM and have at least a workstation inference device.
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...) :)
Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all
#360Earlier quoted context omitted.
it's the frame that's surprisingly - and consistentnly - wrong. You'd think two triangles would be pretty easy to repro; once you get that the rest is easy. It's not like he's asking "draw a pelican on a four-bar linkage suspension mountainbike..."
This is older, but even humans don't have a great concept of how a bicycle works... https://twistedsifter.com/2016/04/artist-asks-people-to-draw...
Wouldn't this be more about being capable of mentally remembering how a bicycle looks versus how it works?
This reminds me of Pictionary. [0] Some people are good and some are really bad.
I am really bad a remembering how items look in my head and fail at drawing in Pictionary. My drawing skills are tied to being able to copy what I see.