Not to mention quantum computing just around the corner with Nvidia’s CUDA Quantum software. The computer hardware has not been developed yet but won’t be long and I’m sure Nvidia’s GPU will be involved along with their software. We all know CUDA is what sets them apart from others

Are you saying it is more of a $1,100 to 1,200 stock

Hey I get frustrated and lash out the FERC thing really did me in I thought everyone was getting their payday then. Oh well live for another day all good here I don’t know the industry but everybody here does keep up the good work and let’s slay these bastards

I already have it but we always want more right

😊

Grouse I don’t know maybe we would be better off if they sold the shell to someone that would use it right away. Just because they have spent over a million to keep it clean doesn’t mean they will use it in the long run. A lot has changed since they acquired 71%

Well now you have to get nasty as I’ve seen you get with others as far as selling I’ll keep my 280,000 shares. As far as giving selling advice you better keep to this blog. Last year you told me to sell my 3,000 shares of Nvidia at $370.00 remember you screen shot some investors letter spewing your know it all crap. If you would have listened about Nvidia you might be in Idaho as we speak tending to your sheep

Not sure what you’re implying by if I really feel that way about if Delfin lied yeah I do but so what not the first company to lie or about they won’t get MARAD or DOE until after election yeah I do if you don’t that’s fine you’re entitled to your opinion. Why is it if you’re not rah rah blah blah a person should sell. I invested in TGLO 6 long years ago because I thought Delfin would RM into it still hope so. Do I have worries that they won’t dam right don’t you or everyone else for that matter. Falling in love with a stock (this not a stock but company Delfin) is the cardinal sin. As far as TGLO being valuable it is very valuable kind of the jest in my last sentence of previous post.

NP what I’m saying is When Delfin had the letter to FERC basically begging for an October certification so they could FID was not truthful. Whether they knew what MARAD was going to do doesn’t matter they knew they needed MARAD to issue to FID and they didn’t have it and didn’t know when they would get it. They haven’t FID without MARAD and now awaiting DOE they won’t. Do you think it was a coincidence Delfin went for 4 years with FERC if FID was imminent. Look I’m not a black helicopter flying over type of guy but I also don’t bury my head in the sand. I personally could care less about Delfin what I do care about is my TGLO investment and at the moment Delfin is what stands between TGLO and success. In my eyes we would all be better off if Delfin sold the shell but they won’t and we will continue to linger.

Delfin’s problems have nothing to do with any administration they are all self inflicted. They knew during FERC they didn’t have MARAD and DOE was coming up and have no plans to FID without them. They have a Senator Cruz ( loud mouth likes to hear his own voice and the last person democrats would listen to) a Italian, British, Japanese company and an American company that won’t SPA write letters, along with some IHubers. Regardless what you think the current administration is not issuing anything before the election period. I invested in TGLO a shell company that currently is being held by Delfin with no immediate plans to use it. That’s the problem wake up, if that’s Debbie Downer so be it.

Their not holding their head at the right angle

You are right it is the amount of Data Centers that are being built and will be built in the near future. Virginia is probably the biggest state with data centers. Of course California won’t have the data centers they can’t charge a car. I think the nasty word nuclear (highly regulated) will end up coming back.

This is a lot of GPU’s


“Meta Spends $30 Billion on a Million NVIDIA GPUs to Train its AI Models
Larger than the Apollo Moon Mission, which cost about $26 billion between 1968 and 1973.


Published on May 3, 2024
by Donna Eva

In a “staggering” revelation, Meta AI chief Yann LeCun confirmed that Meta has obtained $30 billion worth of NVIDIA GPUs to train their AI models. Enough to run a small nation or even put a man on the moon in 1969.

Speaking at the Forging the Future of Business with AI Summit organised by Imagination in Action, LeCun said that more variations of Llama-3 would be out over the next few months, with training and fine-tuning currently taking place.

“Despite all the computers we have on our hands, it still takes a lot of time to fine-tune, but a bunch of variations on those models are going to come out over the next few months,” he said.

Speaking of fine-tuning and training, host John Werner stated that Meta had bought an additional 500,000 GPUs from NVIDIA, taking the total number of NVIDIA GPUs up to a million, with a retail value of $30 billion.

Combining the total costs of the GPUs so far, Werner pointed out that the training of the model exceeded the costs of the entire Apollo space programme, which back in the 1960s, amounted to about $25.4 billion.

Agreeing, LeCun said, “Yeah, it’s staggering, isn’t it? A lot of it, not just training, but deployment, is limited by computational abilities. One of the issues that we’re facing is the supply of GPUs and the cost of them at the moment.

Obviously, adjusted for inflation, the Apollo programme still outsells the Meta in terms of how much was actually spent, with roughly $257 billion spent. But it’s no secret that the cost of GPUs is a continuously growing expense for AI companies.

Recently, OpenAI’s Sam Altman said that he doesn’t care if the company spends upwards of $50 billion a year in developing AGI. The company, as of March, employs as many as 720,000 NVIDIA H100 GPUs for Sora alone. This amounts to about $21.6 billion.

Similarly, all big tech companies are hoping to expand how many GPUs they can obtain by the end of the year, or even by 2025.

Microsoft is aiming for 1.8 million GPUs by the end of the year. Meanwhile, OpenAI hopes to use 10 million GPUs for their latest AI model.

In the meantime, NVIDIA has also been churning out GPUs, with their latest DGX H200 GPU being hand-delivered by CEO Jensen Huang to Altman.

Coming back to LeCun, he pointed out that the need of the hour was the ability to upscale learning algorithms so they could be parallelised across several GPUs. “Progress on this has been kind of slow in the community, so I think we’re kind of waiting for breakthroughs there,” he said.

With that occurring, costs could potentially lower for AI companies, though with increasingly fast upscaling overall, demand could remain the same.”

Was he eaten by cannibals like Joe’s uncle

Yeah I know Meta is using Nvidia I was referring to Apple using Nvidia’s GPU for their generative AI. They will be announcing in the future it would be big if Nvidia was involved

Yeah it would be big news if they used Nvidia’s gear.

In Apples earnings they talked about telling more in the future about their generative ai plans. Does anyone think Nvidia will be in those conversations

Nvidia wants to sell their systems to their customers which also doesn’t help Supermicro

Well then I guess you will get what you deserve then.

I listened to some of AMD earnings call, they increased mi300x yearly sales .5 billion to 4 billion. I think Nvidia loses that much a year, hearing that makes me feel a lot better towards competition. Answers question from article “ Second, as rival semiconductor companies ramp up their efforts to capitalize on the generative artificial intelligence trend.”

Well AMD reports tonight it will be hard to know how Nvidia will react kind of a catch 22 to me. If they have low sales of Mi300x the street could say AI is over (which we know isn’t true). Or if great sales the street will say oh it’s over for Nvidia (also not true). We will see what do you guys think

I see that ford is a partner of Nvidia not sure how intense they are on the autonomous driving aspect but cool beans

That’s okay same question pertains

I always said TGLO would never do anything until Delfin and TGLO were used in the same sentence. I never thought it would be used as a negative but that’s okay if they know about it others will too in a good way

I’m not sure is Ford using Nvidia’s technology

Yes they are a month after probably 95% of the companies, I’m not sure why I guess when they started the company

That would be better

Thanks for the articles I always enjoy them

I think he is a little low on the 50 percent in 2 years good article

The interview was good I guess my expectations were high thinking maybe something big would happen but deep down knowing it wouldn’t. It was good for those people that never heard of Nvidia might say Huh

It will be neat to see Jensen is very good in videos I have seen him in very personal and liked. I especially like him for making us so much money

Not sure 60 minutes following I would think pretty good it’s been on for ever, of course Nvidia is in its quiet period but should be good. A lot of people have never heard of Nvidia but that is about to change

The 60 minutes episodes don’t they usually have 3-5 subject matters so say Jensen will probably be on 10-15 minutes. Does that sound about right I don’t usually watch it.

I agree expecting very positive response as a long I see nothing but long runway

What do you think Nvidia SP will do next week after 60 minutes. 60 minutes is pretty popular and there are a lot of people that don’t know Nvidia. Should be interesting

Well written, I’m not good at writing but I do support it contrary to what people might think.

Article about auto business that will someday become major contributor

Chinese EV maker Xpeng says partnership with Nvidia unaffected by US trade sanctions
Xpeng said local solutions for car chips have emerged despite not being as advanced as imports, but it would keep its options open with domestic suppliers
As of June last year, Nvidia – which is not allowed to ship its advanced AI chips to China – had about a 52 per cent share of the global navigate-on-autopilot market
Thursday, 25 April 2024, 10:00:PM
Chinese electric vehicle (EV) start-up Xpeng said its partnership with Nvidia has not been affected by tighter US trade sanctions, but it will keep its options open with local suppliers.
“We’re very happy with our collaboration with Nvidia and other silicon partners,” Brian Gu Hong-di, vice-chairman and president of Xpeng Motors, said during a press conference on Thursday on the sidelines of the 2024 Beijing Auto Show.
Gu said local solutions for car chips have emerged despite not being as advanced as imports, adding that Xpeng is keeping its options open with domestic suppliers.
Follow and get notified the next time we publish content about
Electric & new energy vehicles
Although Nvidia’s autonomous driving chips are the predominant hardware for supporting assist-driving and more advanced self-driving capabilities developed by carmakers in China, mainland suppliers are quickly catching up amid the country’s drive for self-reliance in semiconductors, especially given that most auto chips do not require cutting-edge manufacturing technologies.
Nissan CEO to respond to EV rivalry in China with bigger line-up, output targets
As of June last year, Nvidia – which is not allowed to ship its advanced AI chips to China due to US trade sanctions – had about a 52 per cent share of the global navigate-on-autopilot market, signing cooperation agreements with 25 automotive original equipment makers globally, according to data from Gasgoo Institute, a Shanghai-based research company.
Nio’s flagship ET5 and ES7 sedans, Li Auto’s L9 model and Xpeng’s G9 all come equipped with Nvidia’s Orin chips. Auto chips are seen by Washington as a less sensitive area of the semiconductor market when it comes to setting trade sanctions.
Help safeguard journalism
Credible news is more vital than ever. Your support helps us protect quality journalism.
Xpeng’s Gu also said that the firm’s smart-driving capabilities for more complex urban traffic environments outshone its competitors. He said the gap between Xpeng and its competitors in city smart driving is “widening”.
“This year, obviously, we want to leverage AI, leveraging large [language] models,” said Gu.
Xpeng has 2,000 engineers working on autonomous driving technologies, ranging from software and architecture to data optimisation, and 3,000 people employed in generative AI.
It will invest 3.5 billion yuan (US$483 million) this year to develop generative AI technologies, which are sweeping every industry from autos and corporate services to consumer electronics.
In 2022, China sold about 7 million smart passenger vehicles equipped with assisted autonomous driving systems, which translates into a market penetration rate of 35 per cent.
In the first half of 2023, that figure rose to almost 43 per cent, according to data from China’s Ministry of Industry and Information Technology.
Autonomous driving companies have been struggling to eke out a profit in China due to heavy costs and low revenue from maintaining large robotaxi fleets.
Baidu, which operates a robotaxi fleet in several cities, said its self-driving unit will focus on generating revenue and profits after burning cash for years.
Smaller players such as WeRide and Pony AI have joined bigger carmakers to monetise software services as they continue to operate driverless robotaxis at a loss.
Xpeng’s Gu paints a realistic outlook, saying it will take longer than five years before robotaxis become a “real commercial operation”.

Excerpts from Googles earnings transcript
CEO

“We offer an industry-leading portfolio of NVIDIA GPUs along with our TPUs. This includes TPU v5p, which is now generally available, and NVIDIA's latest generation of Blackwell GPUs.”

That was the only mention I saw but I didn’t read it just scanned for Nvidia

Hopefully XL175 doesn’t send a comment

lol I forgot got about the toilet flushing