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

321–330 of 400 posts

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

#321

Post by blurbleblurble »

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…

Opus 4.6 was incredible but Opus 4.7 is genuinely frustrating to me so far. It's really sharp but can be so lazy. It's constantly telling me that we should save this for tomorrow, that it's time for bed (in the middle of the day), and very often quite sloppy and bold in its action. These adjustments are getting old. The next crop of open models seems ready to practically replace the big ones as sharp orchestrator agents.

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

#322

Post by unethical_ban »

daemonologist wrote: post #296
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…

First of all nothing you can run locally, on that machine anyways, is going to compare with Opus. (Or even recent Sonnet tbh - some small models benchmark better but fall off a bit in the real world.) This will get you close to like ~Sonnet 4 though: Grab a recent win-vulkan-x64 build of llama.cpp here: https://github.com/ggml-org/llama.cpp/releases - llama.cpp is the engine used by Ollama and common wisdom is to jus…

This is exactly what I have been looking for: Something straight to the point. Thanks a lot!

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

#323

Post by jatins »

senordevnyc wrote: post #165
zozbot234 wrote: post #48

Earlier quoted context omitted.

You need to supervise the model anyway, because you want that code to be long-term maintainable and defect free, and AI is nowhere near strong enough to guarantee that anytime soon. Using the latest Opus for literally everything is just a huge waste of effort.

Yes, but I find supervision much easier and faster with a strong model. It makes fewer dumb mistakes that I have to catch and correct, and it’ll follow my instructions more reliably.

Depends on the task. If it's something that occurs a lot in training data like React/tailwind code then I don't think you need SOTA. Most reasoning models since Sonnet 3.5, Deepseek 3.1 et al will do fine for those tasks.

Doesn't justify 10x the cost in that case imo

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

#324

Post by greyskull »

ihowlatthemoon wrote: post #320
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…

I run a setup similar to yours and I've had the best results with Qwen3.5 27B. Specifically the Q4_K_M variant. https://unsloth.ai/docs/models/qwen3.5 I use llama-server that comes with llama.cpp instead of using ollama. Here are the exact settings I use. llama-server -ngl 99 -c 192072 -fa on --cache-type-k q4_0 --cache-type-v q4_0 --host 0.0.0.0 --sleep-idle-seconds 300 -m Qwen3.5-27B-Q4_K_M.gguf

Thanks, I'll have to continue experimenting. I just ran this model Qwen3.6-35B-A3B-GGUF:UD-Q4_K_XL and it works, but if gemini is to be believed this is saturating too much VRAM to use for chat context.

How did you land on that model? Hard to tell if I should be a) going to 3.5, b) going to fewer parameters, c) going to a different quantization/variant.

I didn't consider those other flags either, cool.

Are you having good luck with any particular harnesses or other tooling?

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

#325

Post by complexworld »

nothinkjustai wrote: post #151
FlyingSnake wrote: post #135

Earlier quoted context omitted.

I tried GLM5.1 last week after reading about it here. It was slow as molasses for routine tasks and I had to switch back to Claude. It also ran out of 5H credit limit faster than Claude.

Z.ai’s cloud offering is poor, try it with a different provider.

could you add some context for why you think it's poor?

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

#327

Post by jxmesth »

embedding-shape wrote: post #198
jxmesth wrote: post #194

Earlier quoted context omitted.

Sorry for the confusion, I was actually talking about their Web based chat. Since most of my work is governance and docs, I just use their Web chats and they just refuse to output proper documents like Claude or Chatgpt do.

Aha... Well, I let Codex (Claude Code would work too) manage/troubleshoot .xlsx files too, seems to handle it just fine (it tends to un-archive them and browse the resulting XML files without issues), seen it do similar stuff for .app and .docx files too so maybe give that a try with other harnesses/models too, they might get it :)

Yeah, it's just way easier to do via the web/mobile app but I'll give using it via the CLI a try. Thanks :)

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

#328

Post by lelanthran »

zozbot234 wrote: post #69
qalmakka wrote: post #65

Earlier quoted context omitted.

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…

If you want to run AI models at scale and with reasonably quick response, there's not many alternatives to datacenter hardware. Consumer hardware is great for repurposing existing "free" compute (including gaming PCs, pro workstations etc. at the higher end) and for basic insurance against rug pulls from the big AI vendors, but increased scale will probably still bring very real benefits.

> If you want to run AI models at scale and with reasonably quick response, there's not many alternatives to datacenter hardware.

Right now, certainly. Things change. What was a datacenter rack yesterday could be a laptop tomorrow.

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

#329

Post by Ey7NFZ3P0nzAe »

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…

In my case, I was also running an ASR model and a TTS model so it was a bit much for my RTX 3090. I opted to offset like 5 layers to the cpu while adding a GPU-only speculative decoding with their 0.8B model.

Working well so far.

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

#330

Post by ulfw »

operatingthetan wrote: post #262
mikenew wrote: post #261

Earlier quoted context omitted.

GLM 5.1 was the model that made me feel like the Chinese models had truly caught up. I cancelled my Claude Max subscription and genuinely have not missed it at all. Some people seem to agree and some don't, but I think that indicates we're just down to your specific domain and usage patterns rather than the SOTA models being objectively better like they clearly used to be.

It seems like people can't even agree which SOTA model is best at any given moment anymore, so yeah I think it's just subjective at this point.

AI is a complete commodity

One model can replace another at any given moment in time.

It's NOT a winner-takes-all industry

and hence none of the lofty valuations make sense.

the AI bubble burst will be epic and make us all poorer. Yay

Post reply on HN