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

Promising results from DeepSeek R1 for code

simonwillison.net

Post reply on HN
Print view View original

11–20 of 765 posts

Re: Promising results from DeepSeek R1 for code

#11

Post by lacedeconstruct »

simonw wrote: post #3

Given these initial results, I'm now experimenting with running DeepSeek-R1-Distill-Qwen-32B for some coding tasks on my laptop via Ollama - their version of that needs about 20GB of RAM on my M2. https://www.ollama.com/library/deepseek-r1:32b It's impressive! I'm finding myself running it against a few hundred lines of code mainly to read its chain of thought - it's good for things like refactoring where it will thi…

The chain of thought is incredibly useful, I almost dont care about the answer now I just follow what I think is interesting from the way it broke the problem down, I tend to get tunnel vision when working for a long time on something so its a great way to revise my work and make sure I am not misunderstanding something

Re: Promising results from DeepSeek R1 for code

#12

Post by pavitheran »

blakesterz wrote: post #4
simonw wrote: post #3

Given these initial results, I'm now experimenting with running DeepSeek-R1-Distill-Qwen-32B for some coding tasks on my laptop via Ollama - their version of that needs about 20GB of RAM on my M2. https://www.ollama.com/library/deepseek-r1:32b It's impressive! I'm finding myself running it against a few hundred lines of code mainly to read its chain of thought - it's good for things like refactoring where it will thi…

Is DeepSeek really that big of a deal that everyone else should worry?

Deepseek is a big deal but we should be happy not worried that our tools are improving.

Re: Promising results from DeepSeek R1 for code

#13

Post by m11a »

m11a wrote: post #6
blakesterz wrote: post #4

Earlier quoted context omitted.

Is DeepSeek really that big of a deal that everyone else should worry?

A lot of the niceness about DeepSeek-R1's usage in coding is that you can see the thought process, which (IME) has been more useful than the final answer. It may well be that o1's chain of thought reasoning trace is also quite good. But they hide it as a trade secret and supposedly ban users for trying to access it, so it's hard to know.

One example from today: I had a coding bug which I asked R1 about. The final answer wasn't correct, but adapting an idea from the CoT trace helped me fix the bug. o1's answer was also incorrect.

Interestingly though, R1 struggled in part because it needed the value of some parameters I didn't provide, and instead it made an incorrect assumption about its value. This was apparent in the CoT trace, but the model didn't mention this in its final answer. If I wasn't able to see the trace, I'd not know what was lacking in my prompt, and how to make the model do better.

I presume OpenAI kept their traces a secret to prevent their competitors from training models with it, but IMO they strategically err'd in doing so. If o1's traces were public, I think the hype around DS-R1 would be relatively less (and maybe more limited to the lower training costs and the MIT license, and not so much its performance and usefulness.)

Re: Promising results from DeepSeek R1 for code

#14

Post by startupsfail »

simonw wrote: post #7
blakesterz wrote: post #4

Earlier quoted context omitted.

Is DeepSeek really that big of a deal that everyone else should worry?

Yeah, it is definitely a big deal. I expect it will be a net positive: they proved that you can both train and run inference against powerful models for way less compute than people had previously expected - and they published enough details that other AI labs are already starting to replicate their results. I think this will mean cheaper, faster, and better models. This FAQ about it is very good: https://stratechery…

This may mean that $3k/task on some benchmarks published by OpenAI are now at slightly lower price tag.

It is possible however that OpenAI was using similar level acceleration in the first place, they’ve just not published the details. And a few engineers left and replicated (or even bested it) in a new lab.

Overall, it’s a good boost, modern software is getting a better fit into new generation of hardware and is performing faster. Maybe we should pay more attention when NVIDIA is publishing their N-times faster ToPS numbers, and not completely dismissing it as marketing.

Re: Promising results from DeepSeek R1 for code

#15

Post by GaggiX »

blakesterz wrote: post #4
simonw wrote: post #3

Given these initial results, I'm now experimenting with running DeepSeek-R1-Distill-Qwen-32B for some coding tasks on my laptop via Ollama - their version of that needs about 20GB of RAM on my M2. https://www.ollama.com/library/deepseek-r1:32b It's impressive! I'm finding myself running it against a few hundred lines of code mainly to read its chain of thought - it's good for things like refactoring where it will thi…

Is DeepSeek really that big of a deal that everyone else should worry?

DeepSeek R1 is o1 but free to use, open source, and also distilled on different models, even the ones that could run on your phone so yeah.

Re: Promising results from DeepSeek R1 for code

#16

Post by netdevphoenix »

simonw wrote: post #7
blakesterz wrote: post #4

Earlier quoted context omitted.

Is DeepSeek really that big of a deal that everyone else should worry?

Yeah, it is definitely a big deal. I expect it will be a net positive: they proved that you can both train and run inference against powerful models for way less compute than people had previously expected - and they published enough details that other AI labs are already starting to replicate their results. I think this will mean cheaper, faster, and better models. This FAQ about it is very good: https://stratechery…

Why did DeepSeek not kept this for themselves? Is this a Meta style scorched earth strategy?

Re: Promising results from DeepSeek R1 for code

#17

Post by csomar »

blakesterz wrote: post #4
simonw wrote: post #3

Given these initial results, I'm now experimenting with running DeepSeek-R1-Distill-Qwen-32B for some coding tasks on my laptop via Ollama - their version of that needs about 20GB of RAM on my M2. https://www.ollama.com/library/deepseek-r1:32b It's impressive! I'm finding myself running it against a few hundred lines of code mainly to read its chain of thought - it's good for things like refactoring where it will thi…

Is DeepSeek really that big of a deal that everyone else should worry?

Everyone else should rejoice. OpenAI is probably cooked, however. Nvidia might be cooked too.

Re: Promising results from DeepSeek R1 for code

#18

Post by the_arun »

lawlessone wrote: post #9
miohtama wrote: post #5

Earlier quoted context omitted.

Also even if the answer is incorrect, you can still cook the eggs on the laptop :)

i spent a months salary on these eggs and can no longer afford to cook them :(

Hey, where are you getting the eggs? I am unable to find them in the market.

Re: Promising results from DeepSeek R1 for code

#19

Post by fibers »

m11a wrote: post #6
blakesterz wrote: post #4

Earlier quoted context omitted.

Is DeepSeek really that big of a deal that everyone else should worry?

A lot of the niceness about DeepSeek-R1's usage in coding is that you can see the thought process, which (IME) has been more useful than the final answer. It may well be that o1's chain of thought reasoning trace is also quite good. But they hide it as a trade secret and supposedly ban users for trying to access it, so it's hard to know.

how many reported cases of banning are there? that sounds insane for asking it to print out its chain of thought

Re: Promising results from DeepSeek R1 for code

#20

Post by coliveira »

blakesterz wrote: post #4
simonw wrote: post #3

Given these initial results, I'm now experimenting with running DeepSeek-R1-Distill-Qwen-32B for some coding tasks on my laptop via Ollama - their version of that needs about 20GB of RAM on my M2. https://www.ollama.com/library/deepseek-r1:32b It's impressive! I'm finding myself running it against a few hundred lines of code mainly to read its chain of thought - it's good for things like refactoring where it will thi…

Is DeepSeek really that big of a deal that everyone else should worry?

It depends on the problem type. If your problem requires math reasoning, deepSeek response is quite impressive and surpasses what most people can do in a single session.
Post reply on HN