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

Qwen3.6-Max-Preview: Smarter, Sharper, Still Evolving

qwen.ai

Post reply on HN
Print view View original

281–290 of 400 posts

Re: Qwen3.6-Max-Preview: Smarter, Sharper, Still Evolving

#281

Post by martinald »

greyskull wrote: post #274

I've been using Claude Code regularly at work for several months, and I successfully used it for a small personal project (a website) not long ago. Last weekend, I explored self-hosting for the first time. Does anyone have a similar experience of having thoroughly used CC/Codex/whatever and also have an analogous self-hosted setup that they're somewhat happy with? I'm struggling a bit. I have 32GB of DDR5 (seems inad…

Try using a MoE model (like Gemma 4 26b-a4b or qwen3.6 35b-a3b) and offload the inference to CPU. If you have enough system RAM (32GB is a bit tight tbh depending on other apps) then this works really well. You may be able to offload some layers to GPU as well though I've had issues with this in MoE models and llama.cpp.

You can keep the KV cache on GPU which means it's pretty damn fast and you should be able to hold a reasonable context window size (on your GPU).

I've had really impressive results locally with this.

I'd strongly recommend cloning llama.cpp locally btw (in wsl2) and asking a frontier model in eg Claude code to set it up for you and tweak it. In my experience the apps that sit on top of llama.cpp don't expose all the options and flags and one wrong flag can mean terrible performance (eg context windows not being cached). If you compile it from source with a coding agent it can look up the actual code when things go wrong.

You should be able to get at least 20-40tok/s on that machine on Gemma 4 which is very usable, probabaly faster on qwen3.6 since it's only 3b active params.

Re: Qwen3.6-Max-Preview: Smarter, Sharper, Still Evolving

#282

Post by ycui1986 »

qalmakka wrote: post #65
visarga wrote: post #41

Earlier quoted context omitted.

I think it is in the interest of chip makers to make sure we all get local models

I think they're in a win-win situation. Big AI companies would love to see local computing die in favour of the cloud because they are well aware the moment an open model that can run on non ludicrous consumer hardware appears, they're screwed. In this situation Nvidia, AMD and the like would be the only ones profiting from it - even though I'm not convinced they'd prefer going back to fighting for B2C while B2B Is s…

There are also many Chines AI-target GPU/NPU producers. You can get a hold of some boards on taobao.com. They are usable in some way.

No, nVidia and AMD are not the only ones benefiting.

Re: Qwen3.6-Max-Preview: Smarter, Sharper, Still Evolving

#283

Post by bink-lynch »

alex7o wrote: post #71

Ok I find it funny that people compare models and are like, opus 4.7 is SOTA and is much better etc, but I have used glm 5.1 (I assume this comes form them training on both opus and codex) for things opus couldn't do and have seen it make better code, haven't tried the qwen max series but I have seen the local 122b model do smarter more correct things based on docs than opus so yes benchmarks are one thing but realit…

I have been using GLM-5.1 with pi.dev through Ollama Cloud for my personal projects and I am very happy with this setup. I use pi.dev with Claude Sonnet/Opus 4.6 at work. Claude Code is great but the latest update has me compacting so much more frequently I could not stand it. I don't miss MCP tool calling when I am using pi.dev; it uses APIs just fine. I actually think GML-5.1 builds better websites than Claude Opus. For my personal projects I am building a full stack development platform and GLM-5.1 is doing a fantastic job.

Re: Qwen3.6-Max-Preview: Smarter, Sharper, Still Evolving

#284

Post by archon810 »

seanw265 wrote: post #215

Kimi K2.6 also released today. I think it's fair to compare the two models. Qwen appears to be much more expensive: - Qwen : $1.3 in / $7.8 out - Kimi : $0.95 in / $4 out -- The announcement posts only share two overlapping benchmark results. Qwen appears to score slightly lower on SWE-Bench Pro and Terminal-Bench 2.0. Qwen : - Teminal-Bench 2.0: 65.4 - SWE-Bench Pro: 57.3 Kimi : - Terminal-Bench 2.0: 66.8 - SWE-Benc…

I wonder if this means a better Cursor Composer model update is coming, since it builds on top of Kimi K2.

Re: Qwen3.6-Max-Preview: Smarter, Sharper, Still Evolving

#285

Post by greyskull »

martinald wrote: post #281
greyskull wrote: post #274

I've been using Claude Code regularly at work for several months, and I successfully used it for a small personal project (a website) not long ago. Last weekend, I explored self-hosting for the first time. Does anyone have a similar experience of having thoroughly used CC/Codex/whatever and also have an analogous self-hosted setup that they're somewhat happy with? I'm struggling a bit. I have 32GB of DDR5 (seems inad…

Try using a MoE model (like Gemma 4 26b-a4b or qwen3.6 35b-a3b) and offload the inference to CPU. If you have enough system RAM (32GB is a bit tight tbh depending on other apps) then this works really well. You may be able to offload some layers to GPU as well though I've had issues with this in MoE models and llama.cpp. You can keep the KV cache on GPU which means it's pretty damn fast and you should be able to hold…

Thanks! These things you're mentioning like "You may be able to offload some layers to GPU...", "You can keep the KV cache on GPU..." configured as part of the llama.cpp? I wouldn't know what to prompt with or how to evaluate "correctness" (outside of literally feeding your comment into claude and seeing what happens).

Aside: what is your tooling setup? Which harness you're using (if any), what's running the inference and where, what runs in WSL vs Windows, etc.

I struggle to even ask the right questions about the workflow and environment.

Re: Qwen3.6-Max-Preview: Smarter, Sharper, Still Evolving

#286

Post by greyskull »

madtowneast wrote: post #277
greyskull wrote: post #274

I've been using Claude Code regularly at work for several months, and I successfully used it for a small personal project (a website) not long ago. Last weekend, I explored self-hosting for the first time. Does anyone have a similar experience of having thoroughly used CC/Codex/whatever and also have an analogous self-hosted setup that they're somewhat happy with? I'm struggling a bit. I have 32GB of DDR5 (seems inad…

You are experiencing the fact that you might not have enough VRAM to load the entire model at a time. You might want to try https://github.com/AlexsJones/llmfit

It's certainly part of the problem. Thanks, I'll give that a shot.

Re: Qwen3.6-Max-Preview: Smarter, Sharper, Still Evolving

#287

Post by lambda »

ezekiel68 wrote: post #201
alex7o wrote: post #71

Ok I find it funny that people compare models and are like, opus 4.7 is SOTA and is much better etc, but I have used glm 5.1 (I assume this comes form them training on both opus and codex) for things opus couldn't do and have seen it make better code, haven't tried the qwen max series but I have seen the local 122b model do smarter more correct things based on docs than opus so yes benchmarks are one thing but realit…

Qwen3-Coder produced much better rust code (that utilized rust's x86-64 vectorized extensions) a few months ago than Claude Opus or Google Gemini could. I was calling it from harnesses such as the Zed editor and trae CLI. I was very impressed.

Their latest, Qwen3.6 35B-A3B is quite capable, and fast and small enough I don't really feel constrained running it locally. Some of the others that I've run that seem reasonably good, like Gemma 4 31B and Qwen3.5 122B-A10B just feel a bit too slow, or OOM my system too often, or run up on cache limits so spend a lot of time re-processing history. But the latest Qwen3.6 is both quite strong, and lightweight enough that it feels usable on consumer hardware.

Re: Qwen3.6-Max-Preview: Smarter, Sharper, Still Evolving

#288

Post by chillfox »

embedding-shape wrote: post #192
jxmesth wrote: post #182

Earlier quoted context omitted.

The only reason I'm stuck with Claude and Chatgpt is because of their tool calling. They do have some pretty useful features like skills etc. I've tried using qwen and deepseek but they can't even output documents. How are you guys handling documents and excels with these tools? I'd love to switch tbh.

> I've tried using qwen and deepseek but they can't even output documents What agent harness did you use? Usually, "write_file", "shell_exec" or similar is two of the first tools you add to an agent harness, after read_file/list_files. If it doesn't have those tools, unsure if you could even call it a agent harness in the first place.

You can make a harness fully functional with just the "shell_exec" tool if you give it access to a linux/unix environment + playwright cli.

Re: Qwen3.6-Max-Preview: Smarter, Sharper, Still Evolving

#289

Post by slopinthebag »

taneq wrote: post #136
OtomotO wrote: post #83

Earlier quoted context omitted.

Many people averted religion (which I can get behind with), but have never removed the dogmatic thinking that lay at its root. As so many things these days: It's a cult. I've used Claude for many months now. Since February I see a stark decline in the work I do with it. I've also tried to use it for GPU programming where it absolutely sucks at, with Sonnet, Opus 4.5 and 4.6 But if you share that sentiment, it's alway…

I wonder to what degree it depends on how easy you find coding in general. I find for the early steps genAI is great to get the ball rolling, but rapidly it becomes more work to explain what it did wrong and how to fix it (and repeat until it does so) than to just fix the code myself.

Yes, this and also taste. What might be perfectly fine for one developer is an abomination for another who can spot the problems with it.

I think in every domain, the better you are the less useful you find AI.

Re: Qwen3.6-Max-Preview: Smarter, Sharper, Still Evolving

#290

Post by bink-lynch »

tasuki wrote: post #235
slopinthebag wrote: post #222

Earlier quoted context omitted.

Opus is about 7 times more expensive than GLM with API pricing. And since you can only use the Opus subscription plan in CC, you're essentially locked into API pricing for Pi and any other harness. So you're either paying $1000's for Opus in Pi, or $30/month for GLM in Pi. If the results are mostly equivalent that's an easy choice for most of us.

Perhaps I'm being extremely daft: If the API is 7 times more expensive, then why is it $1000 vs $30? Or is there a GLM subscription one can use with Pi? Certainly not available in my (arguably outdated) Pi.

I am using GLM-5.1 in pi.dev through Ollama Cloud. I am able to get by on the $20 plan. I use it a lot and the reset is hourly for sessions and weekly overall. This is the first week I got close to the limit before reset at about 85% used. I am probably using it about 4 hours a day on average 6 or 7 days per week.
Post reply on HN