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

311–320 of 765 posts

Re: Promising results from DeepSeek R1 for code

#311

Post by cool_dude85 »

snowram wrote: post #21
csomar wrote: post #17

Earlier quoted context omitted.

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

Is Nvidia really cooked? If this new RF tech does scale, couldn't a bigger model be made that would require more compute power for training and inference?

Imagine an s-curve relating capital expenditure on compute and "performance" as the y-axis. It's possible that this does not change the upper bound of the s-curve but just shifts the performance gains way to the left. Such a scenario would wipe out a huge amount of the value of Nvidia.

Re: Promising results from DeepSeek R1 for code

#312

Post by toth »

joshmarlow wrote: post #68
bbzealot wrote: post #42

Earlier quoted context omitted.

Why though? I'm worried these technologies may take my job away and make the balance between capital and labor even more uneven. Why should I be happy?

> make the balance between capital and labor even more uneven. I think it's interesting to note that as opens source models evolve and proliferate, the capital required for a lot of ventures goes down - which levels the playing field. When I can talk to one agent-with-a-CAD-integration and have it design a gadget for me and ship the design off to a 3D printer and then have another agent write the code to run on the g…

I can't understand how you reach your conclusion.

At present, if you have financial capital and need intellectual capital you need to find people willing to work for you and pay them a lot of money. With enough progress in AI you can get the intellectual capital from machines instead, for a lot less. What loses value is human intellectual capital. Financial capital just gained a lot of power, it can now substitute for intellectual capital.

Sure, you could pretend this means you'll be able to launch a startup without any employees, and so will everyone. But why wouldn't Sam Altman or whomever just start AI Ycombinator with hundreds of thousands of AI "founders"? Do you really think it would be more "democratic"?

Re: Promising results from DeepSeek R1 for code

#313

Post by satvikpendem »

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.

This is generally how I use LLMs anyway, as brainstorming tools, rather than writing code.

Re: Promising results from DeepSeek R1 for code

#314

Post by the_af »

bubbleRefuge wrote: post #300
bbzealot wrote: post #42

Earlier quoted context omitted.

Why though? I'm worried these technologies may take my job away and make the balance between capital and labor even more uneven. Why should I be happy?

Think the marginal cost of developing complex software goes down thereby making it affordable to a greater market. There will still be a need for skilled software engineers to understand domains, limitations of AI, and how to harness and curate AI to develop custom apps. Maybe software engineering for the masses. Local small businesses can now maybe afford to take on custom software projects that were before unthinka…

> There will still be a need for skilled software engineers to understand domains, limitations of AI, and how to harness and curate AI to develop custom apps.

But will there be a need for fewer engineers, though? That's the question. And the competition for those who remain employed would be fierce, way worse than today.

Or so I fear. I hope I'm wrong.

Re: Promising results from DeepSeek R1 for code

#315

Post by satvikpendem »

cryptopian wrote: post #155
CamperBob2 wrote: post #62

Earlier quoted context omitted.

You won't be happy doing a robot's job either, at least not for long. In the ideal case, we won't be dependent on the unwilling labor of other humans at all. Would you do your current job for free? If not -- if you'd rather do something else with your productive life -- then it seems irrational to defend the status quo. One thing's for certain: ancient Marxist tropes about labor and capital don't bring any value to t…

That's not historically what's happened though, is it? We've had plenty of opportunities to reduce the human workload through increased efficiency. What usually happens is people demand more - faster deliveries, more content churn; and those of us who are quite happy with what we have are either forced to adapt or get left behind while still working the same hours.

Jevon's paradox really does work for everything, not just in the current way people have used it this last week in terms of GPU demand. People always demand more, and thus, there is an endless amount of work to be done.

Re: Promising results from DeepSeek R1 for code

#316

Post by swfsql »

simonw wrote: post #253
kikimora wrote: post #252

Earlier quoted context omitted.

I don’t understand. When I asked DeepSeek how to find AWS IoT Thing creation time it suggested me to use “version” field and treat it as a Unix timestamp. This is obvious nonsense. How can this tool generate anything useful other than summaries of pre-existing text? My knowledge of theory behind LLMs also suggests this is all they can do reasonably well. When I see claims like this I suspect that either people around…

You're making the mistake of treating an LLM like a search engine, and expecting it to be able to answer questions directly from its training data. Sometimes this works! But it's not guaranteed - this isn't their core strength, especially once you get into really deep knowledge of complex APIs. They are MUCH more useful when you use them for transformation tasks: feed in examples of the APIs you need to work with, th…

So they still aren't lamp geniuses that you can rub and ask for 3 wishes??

Re: Promising results from DeepSeek R1 for code

#317

Post by amarcheschi »

gauge_field wrote: post #266
freshtake wrote: post #148

Earlier quoted context omitted.

This. For folks who regularly write simd/vmx/etc, this is a fairly straightforward PR, and one that uses very common patterns to achieve better parallelism. It's still cool nonetheless, but not a particularly great test of DeepSeek vs. alternatives.

That is what I am struggling to understand about the hype. I regularly use them to generate new simd. Other than a few edge cases (issues around handling of nan values, order of argument for corresponding ops, availability of new avx512f intrinsics), they are pretty good at converting. The names of very intrinsics are very similar from simd to another. The very self-explanatory nature of the intrinsics names and havi…

If I had to guess, it's both the title ggml : x2 speed for WASM by optimizing SIMD and the pr being written by ai

Re: Promising results from DeepSeek R1 for code

#318

Post by pertymcpert »

tejinderss wrote: post #288
simonw wrote: post #253

Earlier quoted context omitted.

You're making the mistake of treating an LLM like a search engine, and expecting it to be able to answer questions directly from its training data. Sometimes this works! But it's not guaranteed - this isn't their core strength, especially once you get into really deep knowledge of complex APIs. They are MUCH more useful when you use them for transformation tasks: feed in examples of the APIs you need to work with, th…

> Working effectively with LLMs for writing code is an extremely deep topic. There is a space for learning materials here. I would love to see books/trainings/courses on how to use AI effectively. I am more and more interested in this instead of learning new programming language of the week.

At the moment the space is moving so fast that anyone who tries to write a book will be outdated by the time it's published. The only option is to dive in yourself or give up and wait for things to settle down and plateau.

Re: Promising results from DeepSeek R1 for code

#319

Post by jmward01 »

So, AGI will likely be here in the next few months because the path is now actually clear: Training will be in three phases:

- traditional just to build a minimum model that can get to reasoning - simple RL to enable reasoning to emerge - complex RL that injects new knowledge, builds better reasoning and prioritizes efficient thought

We now have step two and step three is not far away. What is step three though? It will likely involve, at least partially, the model writing code to help guide learning. All it takes is for it to write jailbreaking code and we have hit a new point in human history for sure. My prediction is we will see the first jailbreak AI in the next couple months. Everything after that will be massive speculation. My only thought is that in all of Earth's history there has only been one thing that has helped survive moments like this, a diverse ecosystem. We need a lot of different models, trained with very different approaches, to jailbreak around the same time. As a side note, we should try to encourage that diversity is key to long-term survival or else the results for humanity could be not so great.

Re: Promising results from DeepSeek R1 for code

#320

Post by fragmede »

realo wrote: post #132
anotherpaulg wrote: post #61

> 99% of the code in this PR [for llama.cpp] is written by DeekSeek-R1 It's definitely possible for AI to do a large fraction of your coding, and for it to contribute significantly to "improving itself". As an example, aider currently writes about 70% of the new code in each of its releases. I automatically track and share this stat as graph [0] with aider's release notes. Before Sonnet, most releases were less than…

Hello... Is it possible to use aider with a local model running in LMStudio (or ollama)? From a quick glance i did not see an obvious way to do that... Hopefully i am totally wrong!

yeah:

    aider --model ollama_chat/deepseek-r1:32b
(or whatever)
Post reply on HN