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

QwQ-32B: Embracing the Power of Reinforcement Learning

qwenlm.github.io

Post reply on HN
Print view View original

11–20 of 178 posts

Re: QwQ-32B: Embracing the Power of Reinforcement Learning

#11

Post by nazgulsenpai »

gagan2020 wrote: post #5

Chinese strategy is open-source software part and earn on robotics part. And, They are already ahead of everyone in that game. These things are pretty interesting as they are developing. What US will do to retain its power? BTW I am Indian and we are not even in the race as country. :(

If I had to guess, more tariffs and sanctions that increase the competing nation's self-reliance and harm domestic consumers. Perhaps my peabrain just can't comprehend the wisdom of policymakers on the sanctions front, but it just seems like all it does is empower the target long-term.

Re: QwQ-32B: Embracing the Power of Reinforcement Learning

#12

Post by Imnimo »

iamronaldo wrote: post #2

This is insane matching deepseek but 20x smaller?

I wonder if having a big mixture of experts isn't all that valuable for the type of tasks in math and coding benchmarks. Like my intuition is that you need all the extra experts because models store fuzzy knowledge in their feed-forward layers, and having a lot of feed-forward weights lets you store a longer tail of knowledge. Math and coding benchmarks do sometimes require highly specialized knowledge, but if we believe the story that the experts specialize to their own domains, it might be that you only really need a few of them if all you're doing is math and coding. So you can get away with a non-mixture model that's basically just your math-and-coding experts glued together (which comes out to about 32B parameters in R1's case).

Re: QwQ-32B: Embracing the Power of Reinforcement Learning

#13

Post by kmacdough »

7734128 wrote: post #10
iamronaldo wrote: post #2

This is insane matching deepseek but 20x smaller?

Roughly the same number of active parameters as R1 is a mixture-of-experts model. Still extremely impressive, but not unbelievable.

I understand the principles of MOE, but clearly not enough to make full sense of this.

Does each expert within R1 have 37B parameters? If so, is QwQ only truly competing against one expert in this particular benchmark?

Generally I don't think I follow how MOE "selects" a model during training or usage.

Re: QwQ-32B: Embracing the Power of Reinforcement Learning

#16

Post by DiabloD3 »

rspoerri wrote: post #15
bearjaws wrote: post #14

Available on ollama now as well.

i could not find it, where did you?

Ollama's library butchers names, I believe its this: https://ollama.com/library/qwq

The actual name (via HF): https://huggingface.co/Qwen/QwQ-32B

Re: QwQ-32B: Embracing the Power of Reinforcement Learning

#17

Post by Imnimo »

kmacdough wrote: post #13
7734128 wrote: post #10

Earlier quoted context omitted.

Roughly the same number of active parameters as R1 is a mixture-of-experts model. Still extremely impressive, but not unbelievable.

I understand the principles of MOE, but clearly not enough to make full sense of this. Does each expert within R1 have 37B parameters? If so, is QwQ only truly competing against one expert in this particular benchmark? Generally I don't think I follow how MOE "selects" a model during training or usage.

I had a similar confusion previously, so maybe I can help. I used to think that a mixture of experts model meant that you had like 8 separate parallel models, and you would decide at inference time which one to route to. This is not the case, the mixture happens at a much smaller scale.

Instead, the mixture of experts exists within individual layers. Suppose we want to have a big feed-forward layer that takes as input a 1024-element vector, has a hidden size of 8096, and an output size of 1024. We carve up that 8096 hidden layer into 8 1024-sized chunks (this does not have to be the same size as the input). Whenever an input arrives at this layer, a routing function determines which of those 1024-sized chunks should serve as the hidden layer. Every token within a single prompt/response can choose a different chunk when it is processed by this layer, and every layer can have a different routing decision. So if I have 100 layers, each of which has 8 experts, there are 8^100 possible different paths that an individual token could take through the network.

Re: QwQ-32B: Embracing the Power of Reinforcement Learning

#18

Post by bugglebeetle »

gagan2020 wrote: post #5

Chinese strategy is open-source software part and earn on robotics part. And, They are already ahead of everyone in that game. These things are pretty interesting as they are developing. What US will do to retain its power? BTW I am Indian and we are not even in the race as country. :(

Unitree just open-sourced their robot designs:

https://sc.mp/sr30f

China’s strategy is to prevent any one bloc from achieving dominance and cutting off the others, while being the sole locus for the killer combination of industrial capacity + advanced research.

Re: QwQ-32B: Embracing the Power of Reinforcement Learning

#19

Post by mrshu »

DiabloD3 wrote: post #16
rspoerri wrote: post #15

Earlier quoted context omitted.

i could not find it, where did you?

Ollama's library butchers names, I believe its this: https://ollama.com/library/qwq The actual name (via HF): https://huggingface.co/Qwen/QwQ-32B

It indeed seems to be https://ollama.com/library/qwq -- the details at https://ollama.com/library/qwq/blobs/c62ccde5630c confirm the name as "QwQ 32B"

Re: QwQ-32B: Embracing the Power of Reinforcement Learning

#20

Post by asadm »

bugglebeetle wrote: post #18
gagan2020 wrote: post #5

Chinese strategy is open-source software part and earn on robotics part. And, They are already ahead of everyone in that game. These things are pretty interesting as they are developing. What US will do to retain its power? BTW I am Indian and we are not even in the race as country. :(

Unitree just open-sourced their robot designs: https://sc.mp/sr30f China’s strategy is to prevent any one bloc from achieving dominance and cutting off the others, while being the sole locus for the killer combination of industrial capacity + advanced research.

Not really. It seems unitree didn't open source anything. Not anything useful.
Post reply on HN