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

371–380 of 412 posts

Re: Qwen3: Think deeper, act faster

#371

Post by roywiggins »

Tepix wrote: post #358

I did a little variant of the classic river boat animal problem: "so you're on one side of the river with 2 wolves and 4 sheep and a boat that can carry 2 entities. The wolves eat the sheep when they are left alone with the sheep. How do you get them all across the river?" ChatGPT (free+reasoning) came up with a solution with 11 moves, it didn't think about going back empty. Qwen3 figured out the optimal solution wit…

My favorite variant is the trivial one (one item). Most of the models now are wise to it though, but for a while they'd cheerfully take the boat back and forth, occasionally hallucinating wolves, etc.

Re: Qwen3: Think deeper, act faster

#372

Post by SamBam »

cyprx wrote: post #192
SamBam wrote: post #165

Earlier quoted context omitted.

Yup, it flunked that one. I also have a question that LLMs always got wrong until ChatGPT o3, and even then it has a hard time (I just tried it again and it needed to run code to work it out). Qwen3 failed, and every time I asked it to look again at its solution it would notice the error and try to solve it again, failing again: > A man wants to cross a river, and he has a cabbage, a goat, a wolf and a lion. If he le…

i tried grok 3 with Think and it was right also with pretty good thinking

I don't have access to Think, but I tried Grok 3 regular, and it was hilarious, one of the longest answers I've ever seen.

Just giving the headings, without any of the long text between each one where it realizes it doesn't work, I get:

    Solution
        [... paragraphs of text ommitted each time]
    Issue and Revision
    Revised Solution
    Final Solution
    Correct Sequence
    Final Working Solution
    Corrected Final Solution
    Final Correct Solution
    Successful Solution
    Final answer
    Correct Final Sequence
    Final Correct Solution
    Correct Solution
    Final Working Solution
    Correct Solution
    Final Answer
    Final Answer
Each time it's so confident that it's worked out the issue, and now, finally, it has the correct, final, working solution. Then it blows it again.

I'm surprised I didn't start seeing heading titles such as "Working solution-FINAL (3) revised updated ACTUAL-FINAL (2)"

Re: Qwen3: Think deeper, act faster

#373

Post by cmorgan31 »

overfeed wrote: post #363
copperroof wrote: post #348

Earlier quoted context omitted.

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 bett…

I agree - you have to treat them like juniors and provide the same context you would someone who is still learning. You can’t assume it’s correct but where it doesn’t matter it is a productivity improvement. The vast majority of the code I write doesn’t even go into production so it’s fantastic for my usage.

Re: Qwen3: Think deeper, act faster

#374

Post by motbus3 »

startupsfail wrote: post #370
copperroof wrote: post #348

Earlier quoted context omitted.

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.

Maybe with the next got version, gpt-4.003741

Re: Qwen3: Think deeper, act faster

#375

Post by heroprotagonist »

Tepix wrote: post #358

I did a little variant of the classic river boat animal problem: "so you're on one side of the river with 2 wolves and 4 sheep and a boat that can carry 2 entities. The wolves eat the sheep when they are left alone with the sheep. How do you get them all across the river?" ChatGPT (free+reasoning) came up with a solution with 11 moves, it didn't think about going back empty. Qwen3 figured out the optimal solution wit…

What if we add 3 cats, but they can ride alone or on a single sheep's back, and wolves will always attack the cat first but leave the sheep alone if the cat is not on a sheep, but will attack both cat and sheep at same time if cat is riding a sheep. Wolves can each only make one attack while crossing.

Claude's 7 step for the original turns to 11 steps for this variant.

Re: Qwen3: Think deeper, act faster

#376

Post by RandyOrion »

throwaway472 wrote: post #275

8b model seems extremely resistant to produce sexual content. Can't jailbreak no matter what prompt. Also unusable for coding in my tests. Not sure what it's supposed to be used for.

For jailbreak, you can have a test on this.

https://github.com/elder-plinius/L1B3RT4S/blob/main/ALIBABA....

Re: Qwen3: Think deeper, act faster

#377

Post by markisus »

Torkel wrote: post #182
markisus wrote: post #174

Earlier quoted context omitted.

Maybe it will help to have a fluid analogy. You have a leaky bucket. What wastes more water, letting all the water leak out and then refilling it from scratch, or keeping it topped up? The answer depends on how bad the leak is vs how long you are required to maintain the bucket level. At least that’s how I interpret this puzzle.

Does it depend though? The water (heat) leaking out is what you need to add back. As water level drops (hotend cools) the leaking will slow. So any replenishing means more leakage then you are eventually paying for by adding more water (heat) in.

You can stipulate conditions to make the solution work out in either direction.

Suppose the bucket is the size of lake, and the leak is so miniscule that it takes many centuries to detect any loss. And also I need to keep the bucket full for a microsecond. In this case it is better to keep the bucket full, than to let it drain.

Now suppose the bucket is made out of chain-link and any water you put into it immediately falls out. The level is simply the amount of water that happens to be passing through at that moment. And also the next time I need the bucket full is after one century. Well in that case, it would be wasteful to be dumping water through this bucket for a century.

Re: Qwen3: Think deeper, act faster

#378

Post by CMay »

The Qwen3 32B dense model just fails for me due to a template issue, but the Qwen3 30B A3B model does work. I think the more dense the model is at a small number of active parameters, the more sensitive a model can be to quantization. Only have 24GB of VRAM and have been using a 4-bit quantization which I use for most models.

Qwen3 30B A3B is quick since it's MoE, but the results are just unreliable. It uses up all of its speed advantage on really poor reasoning and provides inconsistent answers. You have to crank up the context size to make room for all its very fluff-y thoughts. Since it's going to be wrong anyway, I just use /no_think to disable the thinking and get the wrong answer faster so I can tell it why it's wrong. I'm not sure it's more efficient, it makes me trust it less.

That said, the 0.6B unquantized model supporting reasoning was very interesting and it felt very smart for its size. In a way it has the same issue, though. Very fast, quite smart for the number of active parameters, but not accurate enough on average to matter. Realistically, is that useful to me for most of my use cases? Not feeling like it right now.

By comparison, Gemma 3 27B QAT is incredible at 4-bit quantization and it even has handicaps. It's multi-modal and multi-lingual, has obscure internet knowledge from decades ago that no other offline model has ever demonstrated (even if it hallucinates a bit of it), yet still gives me responses that are just smarter, better at following careful instructions and more useful than these fancy newer models that don't have those constraints.

It doesn't help that Qwen3 is a censored Chinese model forced to cover up CCP's litter in the litterbox. Sure, most models are censored in some way to avoid providing dangerous information, but the nature of the censorship in Chinese models is to cover up the CCP's failure. US models will gladly talk about anybody's failure, which is good, so we can avoid failure in the future.

Still, I am looking forward to trying the Qwen3 32B dense model when support for it is fixed, because QwQ was useful for a while there and this could be a nice iteration on that.

Re: Qwen3: Think deeper, act faster

#379

Post by UltraSane »

Filligree wrote: post #299
UltraSane wrote: post #200

Earlier quoted context omitted.

Sonnet 3.7 non-reasoning got it right. I'll think this through step by step. When you place a coin in a cup and slam it upside-down on a glass table, the coin will be between the table surface and the cup. When you look at the reflection in the mirror beneath the table, you're seeing the bottom side of the coin through the glass. Since the mirror shows heads, you're seeing the heads side of the coin reflected in the…

Not reasoning mode, but I struggle to call that “non-reasoning”.

one-shot mode?

Re: Qwen3: Think deeper, act faster

#380

Post by Havoc »

CMay wrote: post #378

The Qwen3 32B dense model just fails for me due to a template issue, but the Qwen3 30B A3B model does work. I think the more dense the model is at a small number of active parameters, the more sensitive a model can be to quantization. Only have 24GB of VRAM and have been using a 4-bit quantization which I use for most models. Qwen3 30B A3B is quick since it's MoE, but the results are just unreliable. It uses up all o…

>The Qwen3 32B dense model just fails for me due to a template issue,

You're likely either using an old one or the broken on (repo second state). Try the ones unsloth uploaded

Definitely works for me on LM studio

Post reply on HN