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

Nvidia’s $589B DeepSeek rout

finance.yahoo.com

Post reply on HN
Print view View original

521–530 of 1001 posts

Re: Nvidia’s $589B DeepSeek rout

#521

Post by gitfan86 »

TheAlchemist wrote: post #474

Nvidia valuation = $3000B Nvidia profits during AI madness last year = $65B (last 4 quarters) Nvidia profits during normal year = $5B This stock could drop 90% from here, and still be expensive. The numbers are absolutely crazy and make no sense at all. Stargate project is aiming to invest $500B over 4 years. Those $500B are a pipe dream, but let's suppose for a second that all of that $500B will be Nvidia profits an…

Your numbers are wrong. The last quarter profit was $19B and the projected profit is 21B next quarter. That is 84B/year profit with zero growth.

If META,STARGATE,xAI,etc.. all increased spending rapidly you could get to 200B profit rate in 2026.

3T / 200B = 15

That means they could return a 6.6 dividend which is higher than 10 year bonds and is in no way overvalued by historial standards.

All that to say, I sold all my NVDA last month because I don't think Wall St. is buying that everyone is going to actually invest that much.

Re: Nvidia’s $589B DeepSeek rout

#522

Post by jcgrillo »

elihu wrote: post #450
ozten wrote: post #431

NVIDIA sells shovels to the gold rush. One miner (Liang Wenfeng), who has previously purchased at least 10,000 A100 shovels... has a "side project" where they figured out how to dig really well with a shovel and shared their secrets. The gold rush, wether real or a bubble is still there! NVIDA will still sell every shovel they can manufacture, as soon as it is available in inventory. Fortune 100 companies will still…

Jevon's paradox would imply that there's good reason to think that demand for shovels will increase. AI doesn't seem to be one of those things where society as a whole will say, "we have enough of that; we don't need any more". (Many individual people are already saying that, but they aren't the people buying the GPUs for this in the first place. Steam engines weren't universally popular either when they were introdu…

> AI doesn't seem to be one of those things where society as a whole will say, "we have enough of that; we don't need any more".

Really? Has anyone made a useful, commercially successful product with it yet?

Re: Nvidia’s $589B DeepSeek rout

#523

Post by m00x »

KiwiJohnno wrote: post #399
paxys wrote: post #341

IMO this is less about DeepSeek and more that Nvidia is essentially a bubble/meme stock that is divorced from the reality of finance and business. People/institutions who bought on nothing but hype are now panic selling. DeepSeek provided the spark, but that's all that was needed, just like how a vague rumor is enough to cause bank runs.

Not quite, I believe this sell off was caused by DeepSeek showing with their new model that the hardware demands of AI are not necessarily as high as everyone has assumed (as required by competing models). I've tried their 7b model, running locally on a 6gb laptop GPU. Its not fast, but the results I've had have rivaled GPT4. Its impressive.

That's a pretty terrible take.

People who can use the 585B model will use the best model they can have. What DeepSeek really did was start an AI "space race" to AGI with China, and this race is running on Nvidia GPUs.

Some hobbyists will run the smaller model, but if you could, why not use the bigger & better one?

Model distillation has been a thing for over a decade, and LLM distillation has been widespread since 2023 [1].

There is nothing new in being able to leverage a bigger model to enrich smaller models. This is what people that don't understand the AI space got out of it, but it's clearly wrong.

OpenAI has smaller models too with o1 mini and o4 mini, and phi-1 has shown that distillation could make a model 10x smaller perform as well as a much bigger model. The issue with these models is that they can't generalize as well. Bigger models will always win at first, then you can specialize them.

Deepseek also showed that Nvidia GPUs could be more memory-efficient, which catapults Nvidia even further ahead of upcoming processors like Groq or AMD.

[1] https://arxiv.org/abs/2305.02301

Re: Nvidia’s $589B DeepSeek rout

#524

Post by raincole »

TheAlchemist wrote: post #474

Nvidia valuation = $3000B Nvidia profits during AI madness last year = $65B (last 4 quarters) Nvidia profits during normal year = $5B This stock could drop 90% from here, and still be expensive. The numbers are absolutely crazy and make no sense at all. Stargate project is aiming to invest $500B over 4 years. Those $500B are a pipe dream, but let's suppose for a second that all of that $500B will be Nvidia profits an…

> And you know what ? In that scenario Nvidia would STILL be overvalued by historical standards !

It would makes it's PE ratio about 24 (3000/(500/4)).

AAPL's PE is about 36 at this moment.

So no, not "overvalued by historical standards".

Re: Nvidia’s $589B DeepSeek rout

#525

Post by m00x »

zozbot234 wrote: post #375
tgtweak wrote: post #354

Earlier quoted context omitted.

Hype buyers are also Hype sellers - anything Nvidia was last week is exactly what it is this week - DeepSeek doesn't really have any impact on Nvidia sales - Some argument could be made that this can shift compute off of cloud and onto end user devices, but that really seems like a stretch given what I've seen running this locally.

The full DeepSeek model is ~700B params or so - way too large for most end users to run locally. What some folks are running locally is fine-tuned versions of Llama and Qwen, that are not going to be directly comparable in any way.

Many people are missing this due to journalists completely missing the point when presenting these facts.

Distilled models are nothing new.

Re: Nvidia’s $589B DeepSeek rout

#526

Post by int_19h »

brokencode wrote: post #376
DiscourseFan wrote: post #343

Earlier quoted context omitted.

No, because what this implies is that the Chinese have better labor power in the tech-sector than the US, considering how much more efficient this technology is. Which means that even if US companies adopt these practices, the best workers will still be in China, communicating largely in Chinese, building relationships with other Chinese-speaking people purchasing chinese speaking labor. These relationships are alrea…

What a stretch. One Chinese model makes a breakthrough in efficiency and suddenly China has all the best people in the world? What about all the people who invented LLMs and all the necessary hardware here in the US? What about all the models that leapfrog each other in the US every few months? One breakthrough implies that they had a great idea and implemented it well. It doesn’t imply anything more than that.

It's not just one model, though. DeepSeek is the hot story today but Qwen's QwQ also punched above its weight.

Re: Nvidia’s $589B DeepSeek rout

#527

Post by HDThoreaun »

sumeno wrote: post #519
xiphias2 wrote: post #407

Earlier quoted context omitted.

I believe you that it had to do with the selloff, but I believe that efficiency improvements are good news for NVIDIA: each card just got 20x more useful

When your business is selling a large number of cards to giant companies you don't want them to be 20x more useful because then people will buy fewer of them to do the same amount of work

or people do 30x more work and buy 50% more cards

Re: Nvidia’s $589B DeepSeek rout

#528

Post by int_19h »

jaggs wrote: post #441
jdietrich wrote: post #411

Earlier quoted context omitted.

>medium range (e.g. DeepSeek R1x) AI will probably be good enough for most business and user needs Except R1 isn't "medium range" - it's fully competitive with SOTA models at a fraction of the cost. Unless you need multimodal capability or you're desperate to wring out the last percentage point of performance, there's no good reason to use a more expensive model. The real hidden message is that we're still barely get…

Yes absolutely. I guess I meant medium range in terms of dev and running costs. R1 is a premium product at a corner store price. :) People are also forgetting that High-Flyer's ultimate goal is not applications, it's AGI. Hence the open source. They want to accelerate that process out in the open as fast as they can.

I don't think a "good enough" AI is happening for a while. People will see what the new models can do and want the same. So long as improvements are rapid and visible, the demand will keep rising.

Re: Nvidia’s $589B DeepSeek rout

#529

Post by nullc »

Analemma_ wrote: post #371
tgtweak wrote: post #354

Earlier quoted context omitted.

Hype buyers are also Hype sellers - anything Nvidia was last week is exactly what it is this week - DeepSeek doesn't really have any impact on Nvidia sales - Some argument could be made that this can shift compute off of cloud and onto end user devices, but that really seems like a stretch given what I've seen running this locally.

I agree hype is a big portion of it, but if DeepSeek really has found a way to train models just as good as frontier ones for a hundredth of the hardware investment, that is a substantial material difference for Nvidia's future earnings.

I don't see how that follows.

Making training more effective makes every unit of compute spent on training more valuable. This should increase demand unless we've reached a point where better models are not valuable.

The openness of DeepSeek's approach also means that there will be more smaller entities engaging in training rather than a few massive entities that have more ability to set the price they pay.

Plus reasoning models substantially increase inference costs, since for each token of output you may have hundreds of tokens of reasoning.

Arguments on the point can go both ways, but I think on the balance I would expect any improvements in efficiency increase demand.

Re: Nvidia’s $589B DeepSeek rout

#530

Post by datavirtue »

KiwiJohnno wrote: post #399
paxys wrote: post #341

IMO this is less about DeepSeek and more that Nvidia is essentially a bubble/meme stock that is divorced from the reality of finance and business. People/institutions who bought on nothing but hype are now panic selling. DeepSeek provided the spark, but that's all that was needed, just like how a vague rumor is enough to cause bank runs.

Not quite, I believe this sell off was caused by DeepSeek showing with their new model that the hardware demands of AI are not necessarily as high as everyone has assumed (as required by competing models). I've tried their 7b model, running locally on a 6gb laptop GPU. Its not fast, but the results I've had have rivaled GPT4. Its impressive.

Exactly, and firing up reactors to train models just lost all its luster. Those standing before the Stargate will be bored with the whole thing by then end of the week.
Post reply on HN