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

Qwen3: Think deeper, act faster

qwenlm.github.io

Post reply on HN
Print view View original

361–370 of 412 posts

Re: Qwen3: Think deeper, act faster

#361

Post by hangonhn »

janalsncm wrote: post #147
WhitneyLand wrote: post #125

China is doing a great job raising doubt about any lead the major US labs may still have. This is solid progress across the board. The new battlefront may be to take reasoning to the level of abstraction and creativity to handle math problems without a numerical answer (for ex: https://arxiv.org/pdf/2503.21934 ). I suspect that kind of ability will generalize well to other areas and be a significant step toward human…

No kidding. I’ve been playing around with Hunyuan 2.5 that just came out and it’s kind of amazing.

What do you use to run it? Can it be run locally on a Macbook Pro or something like RTX 5070 TI?

Re: Qwen3: Think deeper, act faster

#362

Post by furyofantares »

senordevnyc wrote: post #350
furyofantares wrote: post #145

Earlier quoted context omitted.

I similarly have a small, simple spatial reasoning problem that only reasoning models get right, and not all of them, and which Qwen3 on max reasoning still gets wrong. > I put a coin in a cup and slam it upside-down on a glass table. I can't see the coin because the cup is over it. I slide a mirror under the table and see heads. What will I see if I take the cup (and the mirror) away?

My favorite part of the genre of “questions an LLM still can’t answer because they’re useless!” is all the people sharing results from different LLMs where they clearly answer the question correctly.

I use LLMs extensively and probably should not be bundled into that genre as I've never called LLMs useless.

Re: Qwen3: Think deeper, act faster

#363

Post by overfeed »

copperroof wrote: post #348
jim180 wrote: post #274

Earlier quoted context omitted.

Absolutely. All models ar terrible with Objective-C and Swift, compared to let's say JS/HTML/Python. However, I've realized that Claude Code is extremely useful for generating somewhat simple landing pages for some of my projects. It spits out static html+js which is easy to host, with somewhat good looking design. The code isn't the best and to some extent isn't maintainable by a human at all, but it gets the job do…

I’ve gotten 0 production usable python out of any LLM. Small script to do something trivial, sure. Anything I’m going to have to maintain or debug in the future, not even close. I think there is a _lot_ of terrible python code out there training LLMs, so being a more popular language is not helpful. This era is making transparent how low standards really are.

> I’ve gotten 0 production usable python out of any LLM

Fascinating, I wonder how you use it because once I decompose code to modules and function signatures, Claude[0] is pretty good at implementing Python functions. I'd say it one-shots 60% of the times, I have to tweak the prompt or adjust the proposed diffs 30%, and the remaining 10% is unusable code that I end up writing by hand. Other things Claude is even better at: writing tests, simple refactors within a module, authoring first-draft docstrings, adding context-appropriate type hints.

0. Local LLMs like Gemma3, Qwen-coder seem to be in the same ballpark in terms of capabilities, it's just that they are much slower on my hardware. Except for the 30b Qwen3 MoE that was released a day ago, that one is freakin' fast.

Re: Qwen3: Think deeper, act faster

#364

Post by 85392_school »

Alifatisk wrote: post #344
stavros wrote: post #340

Earlier quoted context omitted.

If OpenAI's flagship model can't add a simple calendar event, that doesn't do much to assuage my disappointment...

I remember vividly that the focus on GPT-4.1 to speak more humane and be more philosophical when speaking. I remember something like that. That model is special and is not meant like a next generation of their other models like 4o and o3. You should try a different model for your task.

I think you're confusing GPT-4.5 with GPT-4.1. GPT-4.1 is their recommended model for non-reasoning API use.

Re: Qwen3: Think deeper, act faster

#366

Post by janalsncm »

hangonhn wrote: post #361
janalsncm wrote: post #147

Earlier quoted context omitted.

No kidding. I’ve been playing around with Hunyuan 2.5 that just came out and it’s kind of amazing.

What do you use to run it? Can it be run locally on a Macbook Pro or something like RTX 5070 TI?

3d.hunyuan.tencent.com

Re: Qwen3: Think deeper, act faster

#367

Post by janalsncm »

Alifatisk wrote: post #235
janalsncm wrote: post #147

Earlier quoted context omitted.

No kidding. I’ve been playing around with Hunyuan 2.5 that just came out and it’s kind of amazing.

Where do you play with it? What shocks you about it? Anything particular?

3d.hunyuan.tencent.com

Re: Qwen3: Think deeper, act faster

#368

Post by RandyOrion »

aurareturn wrote: post #218
RandyOrion wrote: post #201

Earlier quoted context omitted.

YMMV. Parameter efficiency is an important consideration, if not the most important one, for local LLMs because of the hardware constraint. Do you guys really have GPUs with 80GB VRAM or M3 ultra with 512GB rams at home? If I can't run these ultra large MoEs locally, then these models mean nothing to me. I'm not a large LLM inference provider after all. What's more, you also lose the opportunities to fine-tune these…

What people actually want is something like GPT4o/o1 running locally. That's the dream for local LLM people. Running a 7b model for fun is not what people actually want. 7b models are very niche oriented.

For a local LLM, you can't really ask for a certain performance level, it is what it is.

Instead, you can ask for the architecture, be it dense or MoE.

Besides, let's assume the best open weight LLM for now is deepseek r1, is it practical for you to run r1 locally? If not, r1 means nothing to you.

Maybe r1 will be surpassed by llama 4 behemoth. Is it practical for you to run behemoth locally? If not, behemoth also means nothing to you.

Re: Qwen3: Think deeper, act faster

#369

Post by pxc »

Me1000 wrote: post #236
pxc wrote: post #155

Earlier quoted context omitted.

When I tried to reproduce this, DeepSeek refused to answer the question.

There’s an important distinction between the open weight model itself and the deepseek app. The hosted model has a filter, the open weight does not.

I didn't know that! That gives me another reason to play with it at home. Thanks for cluing me in. :)

Re: Qwen3: Think deeper, act faster

#370

Post by startupsfail »

copperroof wrote: post #348
jim180 wrote: post #274

Earlier quoted context omitted.

Absolutely. All models ar terrible with Objective-C and Swift, compared to let's say JS/HTML/Python. However, I've realized that Claude Code is extremely useful for generating somewhat simple landing pages for some of my projects. It spits out static html+js which is easy to host, with somewhat good looking design. The code isn't the best and to some extent isn't maintainable by a human at all, but it gets the job do…

I’ve gotten 0 production usable python out of any LLM. Small script to do something trivial, sure. Anything I’m going to have to maintain or debug in the future, not even close. I think there is a _lot_ of terrible python code out there training LLMs, so being a more popular language is not helpful. This era is making transparent how low standards really are.

Try o4-mini-high. It’s getting there.
Post reply on HN