[flagged]
Qwen3.6-35B-A3B: Agentic coding power, now open to all
541–550 of 563 posts
Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all
#542I have a Macbook M3 Max with 128GB of RAM.
How close to Opus 4.6 can I get with this? Realistic, real-world usage. And I mean not sitting there for minutes waiting the model to finish saying hello, or being able to use it for anything more than a pelican riding a bicycle.
I'm asking because I'm always seeing excited replies, then I get excited, then I spend minutes to hours setting up the model and then, after first use I forget it exists for one reason or another.
Can I get any realistic use out of this?
Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all
#543I have a Macbook M3 Max with 128GB of RAM. How close to Opus 4.6 can I get with this? Realistic, real-world usage. And I mean not sitting there for minutes waiting the model to finish saying hello, or being able to use it for anything more than a pelican riding a bicycle. I'm asking because I'm always seeing excited replies, then I get excited, then I spend minutes to hours setting up the model and then, after first…
You'd be the best person in this thread to answer this question.
Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all
#544Just tried to use qwen3.6-35b-a3b-bf16 + omlx running a pi session to use my HN cli to do a sentiment analysis on this story and opus4.7 story. I’m getting ~40tk/s on a M3 Ultra Mac Studio and the tool use consistency has been held up well. Even when passing 100k tokens, the session was still going strong. Here is the full sentiment analysis report it produced:
https://gist.github.com/duh17/2db5351da026cec4bd4f46e169e75e...
Here is the full session:
https://pi.dev/session/#c3d003becb1bfcc7ffbca04e89e1adf8
This is by far my smoothest agentic session using a local model of any size. The output quality and speed has really struct the right balance. Very impressive release
Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all
#545I have a Macbook M3 Max with 128GB of RAM. How close to Opus 4.6 can I get with this? Realistic, real-world usage. And I mean not sitting there for minutes waiting the model to finish saying hello, or being able to use it for anything more than a pelican riding a bicycle. I'm asking because I'm always seeing excited replies, then I get excited, then I spend minutes to hours setting up the model and then, after first…
It won’t be a fair comparison against opus-4.6 but it will run quite well on your machine. I’ve tested qwen3.5 27B, Gemma4, minimax2.5 and Glm4.7 before on my m3 ultra. And i’d say this is the first model that I’m able to use for full agentic sessions. here is a pi session i just did and it worked quite well surprisingly: https://pi.dev/session/#c3d003becb1bfcc7ffbca04e89e1adf8
Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all
#546Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all
#547I have a Macbook M3 Max with 128GB of RAM. How close to Opus 4.6 can I get with this? Realistic, real-world usage. And I mean not sitting there for minutes waiting the model to finish saying hello, or being able to use it for anything more than a pelican riding a bicycle. I'm asking because I'm always seeing excited replies, then I get excited, then I spend minutes to hours setting up the model and then, after first…
It won’t be a fair comparison against opus-4.6 but it will run quite well on your machine. I’ve tested qwen3.5 27B, Gemma4, minimax2.5 and Glm4.7 before on my m3 ultra. And i’d say this is the first model that I’m able to use for full agentic sessions. here is a pi session i just did and it worked quite well surprisingly: https://pi.dev/session/#c3d003becb1bfcc7ffbca04e89e1adf8
Thank you! That actually looks quite impressive.
What seems very promising is that thinking blocks look coherent for the lack of a better word, and not that far away from thinking blocks (or rather, summaries) that I see from Claude models.
I think this could actually work for targeted worker agents that get explicit, detailed task instructions from better models.
I'll be trying this tomorrow in my workflow.
Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all
#548I 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 doi…
Now that I have tried out on a few tasks, Qwen3.6 is a huge jump in capability. It can make improvements to a project that qwen3.5 always struggled with.
Could you share more about your config? I've also got a 3060 12gb and 64gb of ram, but I've never got local models running well enough to be useful
Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all
#549Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all
#550Earlier quoted context omitted.
Fwiw, with its predecessor's Qwen3.5-35B-A3B-Q6_K.gguf, on a laptop's 6 GB VRAM and 32 GB RAM, with default llama.cpp settings, I get 20 t/s generation.
That is pretty solid, I have a 2070 with 8GB VRAM and 64GB RAM, but I haven't run too much. I regret not getting a 3090 back when I built this machine.
Nod. Mine was VR dev leftovers. Fwiw, running 6ish prompts in parallel, roughly doubles my aggregate t/s (but requires cooling kludgery). If one's goal is not local, but rather real-time or consistent or transparent or scalable, there's AWS.