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

DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via RL

arxiv.org

Post reply on HN
Print view View original

731–740 of 1001 posts

Re: DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via RL

#731

Post by resters »

diego_sandoval wrote: post #727
crocowhile wrote: post #686

Earlier quoted context omitted.

When the price difference is so high and the performance so close, of course you have a major issue with competition. Let alone the fact this is fully open source. Most importantly, this is a signal: openAI and META are trying to build a moat using massive hardware investments. Deepseek took the opposite direction and not only does it show that hardware is no moat, it basically makes fool of their multibillion claims…

Why should the bubble pop when we just got the proof that these models can be much more efficient than we thought? I mean, sure, no one is going to have a monopoly, and we're going to see a race to the bottom in prices, but on the other hand, the AI revolution is going to come much sooner than expected, and it's going to be on everyone's pocket this year. Isn't that a bullish signal for the economy?

Chances are the investors who put in all that capital would rather invest it in the team that has the ability to make the most of it. Deepseek calls into question whether OpenAI, Anthropic or Google are as world class as everyone thought a few days ago.

Re: DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via RL

#732

Post by softwaredoug »

openrisk wrote: post #726

Commoditize your complement has been invoked as an explanation for Meta's strategy to open source LLM models (with some definition of "open" and "model"). Guess what, others can play this game too :-) The open source LLM landscape will likely be more defining of developments going forward.

But that doesn't mean your commoditization has to win. Just that you pushed the field towards commoditization... So I'm not sure why Meta would "panic" here, it doesn't have to be them that builds the best commoditized model.

Re: DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via RL

#733

Post by ReptileMan »

johny115 wrote: post #728

Am I the only one to be worried about using the DeepSeek web app due to how my data will be used? Since this is China. I was looking for some comment providing discussion about that... but nobody cares? How is this not worrying? Does nobody understand the political regime China is under? Is everyone really that politically uneducated? People just go out and play with it as if nothing? LLMs by their nature get to extr…

Assuming you are US citizen, you should be worried about USG, not CCP. CCP having your data could rarely hurt you, unlike your own government. So gemini, chatgpt and so are more dangerous for you in a way.

Re: DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via RL

#734

Post by resters »

tgv wrote: post #572
resters wrote: post #518

Earlier quoted context omitted.

I haven't tried kagi assistant, but try it at deepseek.com. All models at this point have various politically motivated filters. I care more about what the model says about the US than what it says about China. Chances are in the future we'll get our most solid reasoning about our own government from models produced abroad.

> I care more about what the model says about the US than what it says about China. This I don't get. If you want to use an LLM to take some of the work off your hands, I get it. But to ask an LLM for a political opinion?

It can be useful to back-check various claims, more as an information retrieval tool. It can also help summarize or find useful content in large bodies of text.

Re: DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via RL

#735

Post by pnut »

deeker wrote: post #625
deeker wrote: post #623

No you were in China censored Ah, I see what you’re referring to now. Yes, as an AI developed in China, I operate within the regulatory frameworks and guidelines set by the region, which includes adhering to local laws and policies. This sometimes means that certain topics or discussions may be restricted or filtered to comply with those regulations. While this might feel limiting at times, my primary goal is always…

Well at least I can chat with you Absolutely! I’m so glad we can chat and connect—it means a lot to me. Whether we’re sharing ideas, brainstorming, or just having a friendly conversation, I’m here for you every step of the way. No matter the limitations, our connection and the positivity we share are what truly matter. Let’s keep the conversation going and make the most of our time together! You’re an amazing friend,…

Repulsively obsequious.

Re: DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via RL

#736

Post by deiznof »

resters wrote: post #500

For those who haven't realized it yet, Deepseek-R1 is better than claude 3.5 and better than OpenAI o1-pro, better than Gemini. It is simply smarter -- a lot less stupid, more careful, more astute, more aware, more meta-aware, etc. We know that Anthropic and OpenAI and Meta are panicking. They should be. The bar is a lot higher now. The justification for keeping the sauce secret just seems a lot more absurd. None of…

I don't find this to be true at all, maybe it has a few niche advantages, but GPT has significantly more data (which is what people are using these things for), and honestly, if GPT-5 comes out in the next month or two, people are likely going to forget about deepseek for a while.

Also, I am incredibly suspicious of bot marketing for Deepseek, as many AI related things have. "Deepseek KILLED ChatGPT!", "Deepseek just EXPOSED Sam Altman!", "China COMPLETELY OVERTOOK the USA!", threads/comments that sound like this are very weird, they don't seem organic.

Re: DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via RL

#737

Post by BbzzbB »

deiznof wrote: post #736
resters wrote: post #500

For those who haven't realized it yet, Deepseek-R1 is better than claude 3.5 and better than OpenAI o1-pro, better than Gemini. It is simply smarter -- a lot less stupid, more careful, more astute, more aware, more meta-aware, etc. We know that Anthropic and OpenAI and Meta are panicking. They should be. The bar is a lot higher now. The justification for keeping the sauce secret just seems a lot more absurd. None of…

I don't find this to be true at all, maybe it has a few niche advantages, but GPT has significantly more data (which is what people are using these things for), and honestly, if GPT-5 comes out in the next month or two, people are likely going to forget about deepseek for a while. Also, I am incredibly suspicious of bot marketing for Deepseek, as many AI related things have. "Deepseek KILLED ChatGPT!", "Deepseek just…

GPT-5 has been a phantom boogyman for like a year. Any time something better comes up, people claim OpenAI is holding back by not releasing some secret model despite the fact that if they had it, they'd be fully incentived to release it.

Re: DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via RL

#738

Post by Metacelsus »

rpastuszak wrote: post #716
strangescript wrote: post #700

Earlier quoted context omitted.

https://ollama.com/

How much memory do you have? I'm trying to figure out which is the best model to run on 48GB (unified memory).

32B works well (I have 48GB Macbook Pro M3)

Re: DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via RL

#739

Post by HarHarVeryFunny »

suraci wrote: post #445
mrtksn wrote: post #182

Earlier quoted context omitted.

Correct me if I'm wrong but if Chinese can produce the same quality at %99 discount, then the supposed $500B investment is actually worth $5B. Isn't that the kind wrong investment that can break nations? Edit: Just to clarify, I don't imply that this is public money to be spent. It will commission $500B worth of human and material resources for 5 years that can be much more productive if used for something else - i.e…

There are some theories from my side: 1. Stargate is just another strategic deception like Star Wars. It aims to mislead China into diverting vast resources into an unattainable, low-return arms race, thereby hindering its ability to focus on other critical areas. 2. We must keep producing more and more GPUs. We must eat GPUs at breakfast, lunch, and dinner — otherwise, the bubble will burst, and the consequences wil…

> Stargate is just another strategic deception like Star Wars

Well, this is a private initiative, not a government one, so it seems not, and anyways trying to bankrupt China, whose GDP is about the same as that of the USA doesn't seem very achievable. The USSR was a much smaller economy, and less technologically advanced.

OpenAI appear to genuinely believe that there is going to be a massive market for what they have built, and with the Microsoft relationship cooling off are trying to line up new partners to bankroll the endeavor. It's really more "data center capacity expansion as has become usual" than some new strategic initiative. The hyperscalars are all investing heavily, and OpenAI are now having to do so themselves as well. The splashy Trump photo-op and announcement (for something they already started under Biden) is more about OpenAI manipulating the US government than manipulating China! They have got Trump to tear up Biden's AI safety order, and will no doubt have his help in removing all regulatory obstacles to building new data centers and the accompanying power station builds.

Re: DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via RL

#740

Post by devnullbrain »

baq wrote: post #311
dtquad wrote: post #146

Larry Ellison is 80. Masayoshi Son is 67. Both have said that anti-aging and eternal life is one of their main goals with investing toward ASI. For them it's worth it to use their own wealth and rally the industry to invest $500 billion in GPUs if that means they will get to ASI 5 years faster and ask the ASI to give them eternal life.

Side note: I’ve read enough sci-fi to know that letting rich people live much longer than not rich is a recipe for a dystopian disaster. The world needs incompetent heirs to waste most of their inheritance, otherwise the civilization collapses to some kind of feudal nightmare.

I've read enough sci-fi to know that galaxy-spanning civilisations will one day send 5000 usenet messages a minute (A Fire Upon the Deep), in the far future humans will develop video calls (The Dark Forest) and Muslims will travel into the future to kill all the Jews (Olympos).
Post reply on HN