GPU Economics

“RightScale… they quantified it saying 76% of the respondents complained managing cloud cost is a big problem… And on the other side, you have 85% of server capacity across the globe that’s not used. I think Stanford did a study in 2015… they estimated that about 10 million servers are in [a] comatose state… Not touched for six months.” — 00:41:31

“What we do is we tap into slack server capacity that’s underutilized — I mean, eighty-five percent of capacity out there today sits underutilized in most of the servers, and our job is to bring them, aggregate, and provide an interface.” — 00:18:16

“I wanted to solve the cost problem by tapping into slack capacity — the 85% of server capacity today sits unused. Extreme example would be Intuit… they run at 97 or so percent capacity; come April 16th, they drop it down to three or four percent.” — 00:09:34

“Some of our customers are paying huge amounts to the cloud today, and we believe we can bring that price down by 90% by creating this decentralized economy. Right now with our supply we can guarantee a 50% reduction in price; with our current ecosystem we will try to get to 90% with higher demand.” — 00:22:38

“There is a ton of capacity sitting in data centers. About 85% of the capacity sitting in 4.8 million data centers is not used… there’s this incredible capacity that’s not being used. And there is this, you know, a few companies that are capitalizing on knowledge. And we felt that it was fundamentally broken.” — 00:06:06

“Every attempt to price compute has ended up in an utter failure, EOS being I think a big example… Every time you commoditize something, turns out… there’s always going to be room for people to cheat… compute is very, very hard to commoditize. Instead of trying to commoditize, [we] create a free market that’s driven by auction… it’s like diamonds and it’s not like gold.” — 00:37:34

“It really comes down to price performance. So when you have a price performance metric, where the cost is insignificant, latency becomes insignificant as well… when you do batch optimality… the requirement really is driven by cost. So our thesis is the cost is exponentially lower, about like eight times, nine times lower, latency becomes less important for batch optimal workloads.” — 00:18:21

“Machine learning happens to be our biggest use case, even though Akash is generic compute… because of the cost advantage that we’re seeing with machine learning applications.” — 00:48:56

“Usually we’re looking at north of 85% of capacity that sits idle in these data centers… a lot of them happen to be like GPU clusters as well… they use these massive clusters to design cars… but they only use them like two hours tops a day… All in all, there are about 8.2 million data centers in the world with excess capacity that is just sitting idle.” — 00:06:02

“You’re getting high quality, really high quality bare metal servers for next to nothing. I hope these prices will stay the way they are, but I think right off the bat we’re looking for like two to three x lower than the market… this is bare metal, so there’s nothing in the middle — you’re just getting raw performance that is not possible on the cloud today.” — 00:17:04

“You’re getting high quality, really high quality bare metal servers for next to nothing. I hope these prices will stay the way they are, but I think right off the bat we’re looking for like two to three x lower than the market… this is bare metal, so there’s nothing in the middle — you’re just getting raw performance that is not possible on the cloud today.” — 00:17:43

“Today to run an AI company on the cloud, it’s literally paying 20% of your margins to Amazon… you have this enormous amount of capacity… your utilization rates are significantly low in your data center, north of 85 or so in our conservative estimates — you have underutilized capacity.” — 00:08:29

“The most attractive area for us to focus is machine learning… because that segment is currently suffering with the immense cost problem… [an Andreessen Horowitz] report recently… stated essentially 20 [percent] of margins tend to go to the cloud. It’s extremely hard being a growth stage AI company that’s trying to survive in this increasingly prohibitive layer called cloud.” — 00:05:43

“The most attractive area for us to focus is machine learning… because that segment is currently suffering with the immense cost problem… [an Andreessen Horowitz] report recently… stated essentially 20 [percent] of margins tend to go to the cloud. It’s extremely hard being a growth stage AI company that’s trying to survive in this increasingly prohibitive layer called cloud.” — 00:05:43

“The model that we envisioned for an AI company to function is more of a peer-to-peer cloud… they have a server from the offices that has GPU clusters… they only use these clusters maybe two or three hours a day while they’re training these models… they can install Akash and offer this unused capacity and earn credits in exchange… and they can use these credits for when they want to use peak capacity. That’s really when you see providers becoming users and users becoming providers.” — 00:05:43

“The model that we envisioned for an AI company to function is more of a peer-to-peer cloud… they have a server from the offices that has GPU clusters… they only use these clusters maybe two or three hours a day while they’re training these models… they can install Akash and offer this unused capacity and earn credits in exchange… and they can use these credits for when they want to use peak capacity. That’s really when you see providers becoming users and users becoming providers.” — 00:05:43

“There’s an article by [Andreessen] Horowitz that stated about 20% of margins for a lot of AI companies go to paying the cloud bills, and… it’s rampant, it’s very very expensive.” — 01:17:44

“Especially as we see a lot of these proof-of-work coins start to switch to proof-of-stake… you’re gonna have a lot of GPUs on the market that… can do something better, and so I’m hoping to see that Akash will be there and ready to fill in those boots and take in all that capacity.” — 01:21:20

“They can install Akash and earn compute credits in AKT tokens in their downtime, and when they want to peak they can use this AKT tokens to actually use for compute. So you essentially create a very amazing mechanism where the provider becomes a tenant and vice versa. It’s a self-serving ecosystem of a peer-to-peer cloud… that’s only possible on Akash.” — 01:19:10

“20% of revenues for these machine learning companies is going to the cloud… We’d love for these companies to install Akash and earn the tokens when they’re not using [their servers], and when they want to peak they can use these tokens to actually get compute. This is the use case where the provider becomes a tenant and vice versa, creating a truly peer-to-peer cloud.” — 00:46:37

“We’re working on futures as well — it’s still early design — I think opening up, creating a futures market for compute will add efficiency. Futures generally add efficiency to pricing.” — 01:05:57

“They could actually install Akash in their clusters and sell the underutilized… capacity back into the network like a power grid model, and earn tokens for that unused capacity, and when they want to use at peak… they can use these tokens… so in this case the provider itself becomes a payer, and that’s really an ideal peer-to-peer GPU cloud that we really want to target.” — 00:38:14

“We have amazing opportunities with Ethereum network… coming off proof of work and going to proof of stake — that opens up a whole range of GPUs on the market, so that’s significantly helpful for AI workloads.” — 00:14:49

“As we were deploying this Kubernetes in data centers across the walls, what we discovered was 85% or so of compute capacity that sits in these data centers remains unused. The reason being most data centers are planned for peak capacity to handle peak loads.” — 00:04:59

“We’ll be launching the world’s first GPU marketplace. So that’s going to be very exciting because there’s an enormous amount of GPUs that are getting essentially unlocked with Ethereum and a lot of these networks going to proof of stake… currently GPUs are extremely expensive on Amazon.” — 00:28:26

“When Ethereum moves to a proof of stake… all that hardware becomes useless. And it’s really good hardware… there is no marketplace, there’s no sort of like liquidity mechanism… So the solution is to install Akash and have users run TensorFlow applications or any of these machine learning applications and take advantage of this chip compute.” — 00:29:14

“Эфир переходит на Proof of Stake, а значит, огромная инфраструктура графических адаптеров, которая поддерживает эфир, станет никому не нужной. Это предоставляет огромную возможность для рынка ИИ, искусственного интеллекта. Мы строим инфраструктуру, ориентированную именно на этот рынок.” (“Ethereum is moving to Proof of Stake, which means the enormous GPU infrastructure supporting Ethereum will no longer be needed by anyone. This presents a huge opportunity for the AI market. We are building infrastructure aimed precisely at that market.”)” — 00:15:47

“Есть довольно много неиспользуемых графических адаптеров по всему миру, и эта цифра будет продолжать расти, потому что многие из этих сетей сейчас переходят от Proof-of-Work на Proof-of-Stake… Когда произойдет переход, мы хотим быть готовы к работе с рынком графических процессоров. Мы считаем, что сможем купить их за полцены, а для рынка искусственного интеллекта это очень важно.” (“There are quite a lot of unused GPUs around the world, and that number will keep growing because many of these networks are moving from Proof-of-Work to Proof-of-Stake… When the transition happens, we want to be ready to work with the GPU market. We believe we’ll be able to buy them at half price, and for the artificial intelligence market that is very important.”)” — 00:30:52

“Ethereum network moving off proof of work and moving on to proof of stake implies that this entire GPU infrastructure Ethereum miners have built is going to become obsolete, and we want to tap into that… and enable a multi-tenancy GPU marketplace… I think it’s going to be huge for us.” — 00:38:02

“Today on the cloud, if you are a machine learning company using Amazon, 20% of your margins go to Amazon, and you and I, even though we don’t run machine learning workloads, are actually paying an invisible Amazon tax.” — 00:38:02

“We leverage essentially 85 or so [percent] underutilized cloud capacity in about 8.4 million data centers to enable anyone with a computer to buy and sell cloud compute in a safe and frictionless market… the tenant… experiences anywhere from 8x to 10x lower cost than what you would pay Amazon.” — 00:00:43

“Later this year we’re coming up with GPU marketplace — it’ll be the world’s first open marketplace for GPU compute, and it’s a big deal because the GPU segment is suffering right now… 20 percent of their margins go to Amazon, and we have enormous capacity, enormous GPU demand, where some of the customers are spending about a million to two million dollars a month on GPU right now — they want to come to Akash.” — 00:17:02

“A16z wrote an article recently that basically said 20% of margins for AI companies goes to Amazon. So it is so big that it is actually hurting their viability to exist as a business. It’s no longer a secondary cost, it is a primary cost, sometimes much more than payroll… I’m particularly excited about AI workloads.” — 00:09:55

“Another interesting event that is going to happen over the next couple of years is Ethereum moving off proof of work and coming to proof of stake — [that] unlocks enormous amount of computing… there’s enormous amount of GPUs that are being unlocked and we want to offer that to a customer base that’s really suffering quite a bit right now.” — 00:28:29

“If you think about Ethereum network moving off proof of work and moving on to proof of stake implies that this entire GPU infrastructure Ethereum miners have built is going to become obsolete, and we want to tap into that… and enable a multi-tenancy GPU marketplace… I think it’s going to be huge for us.” — 00:33:26

“Today on the cloud, if you are a machine learning company using Amazon, 20% of your margins go to Amazon, and you and I, even though we don’t run machine learning workloads, are actually paying an invisible Amazon tax.” — 00:32:42

“When it comes to Akash, the cost advantage of using Akash versus a centralized cloud is one-third… sometimes it is one-fourth, like when compared to Google. And the reason for that is because open marketplaces generally tend to optimize for the forces of demand-supply for pricing, versus an individual setting the price.” — 00:08:32

“It turns out 85 to 90 [percent] of data centers in the world are not used or underutilized… the result is the cost advantage you get on Akash is about one-third than Amazon. So if you’re paying a dollar on Amazon, you pay 30 cents in Akash — that’s just a pure free market at its best.” — 00:05:00

“That should be out in another six months or so if everything goes right — and turns out making a multi-tenancy marketplace for GPUs is not the easiest thing… if you can run TensorFlow on Akash — you can right now, but with GPUs — I think that’s going to be a holy grail.” — 00:15:31

“A big feature that’s upcoming is GPUs. For us, GPUs are one of the most exciting things for me, because today the biggest cost for AI is GPUs. Andreessen Horowitz even wrote an amazing article recently looking at their portfolio companies… this GPU cost, which is about 20% of margins — so 20% of the money, either investment or revenue, these AI companies make goes to Amazon, which is a lot.” — 00:13:25

“Supermini will become a reality when Akash enables GPU, which is in the next few months essentially.” — 00:48:24

“While we were deploying, we discovered that most capacity in these data centers — somewhere north of 85 percent of capacity — remains unused for various reasons, and that’s when the idea for a marketplace came about.” — 00:02:15

“Right now Akash supports CPUs only, but in about six months’ time you’ll start supporting GPUs — the work is underway. What that means is a lower cost GPU-based compute. As you know, GPU is very, very expensive on the cloud, so any cost reduction I think is a big, big win for the data science community.” — 00:03:36

“Machine learning is so expensive that if you look at any S1 filings from web 2 companies like Snowflake or whatnot, there’s a strong indication that 50 percent or so of margins actually go to pay Amazon bills.” — 00:29:07

“The important part of this model is that developers set the price, versus the traditional model where providers set the price… today Akash costs anywhere from half to one-fifth what you would pay on Amazon.” — 00:16:16

“We’re going to have machine learning in another six months hopefully… Ethereum is coming off proof of work to proof of stake, and that’s going to unleash a humongous amount of GPU systems that were primarily used for mining… having machine learning run on these GPU instances at 20% lower cost is going to be a huge win, and we have enterprise customers lined up.” — 00:29:49

“Akash right now offers a price point that’s one-third to one-fifth lower than Amazon. So when you see that sort of arbitrage opportunity… the best way to realize that opportunity is to mine. So we’ve seen quite a bit of miners… and we’re seeing quite a bit of Chia miners… so far it’s been more supply than demand, surprisingly.” — 00:15:32

“There’s a whole rework happening on economics now for Akash. We’re looking at enabling more DeFi primitives into the dCloud, in the sense we want to do futures for compute — a futures market for compute is something that doesn’t exist today and it can be great to do it in an open way.” — 00:32:36

“A big source of GPUs in the next six months is going to be from Ethereum miners. As you know, Ethereum is moving off proof of work to proof of stake — that unlocks [an] enormous amount of GPU capability in the market that has really nowhere to go, and Akash being a prime cloud partner would be an amazing use for us to use these GPUs and offer that at a discount that’s significant from what’s being offered in the market right now.” — 00:26:23

“Immediately after that we’re going to have GPUs — we’re suspecting another three to five months for the GPUs to be ready.” — 01:19:58

“Absolutely. With the GPUs coming, all the Ethereum miners can run Akash on them and repurpose their hardware, and that’s a big bet we’re making actually. So we’re very excited for GPUs.” — 00:50:42

“Akash is a decentralized cloud infrastructure network, essentially a marketplace for cloud-grade compute — offers compute in an open, in a permissionless way, at a price point that’s market driven, usually anywhere from one-fifth to one-third to what you would pay on the market.” — 00:01:57

“GPUs — we are spec ready. We have about 20 customers that are ready to jump on with GPU… it’s a very complicated feature, especially doing generically schedulable GPUs. No one has done GPUs. Even the people that claim that they have GPUs — most GPUs are only used for mining, and that’s a completely different use case. So designing a general purpose GPU cloud is extremely challenging.” — 00:23:26

“Currently GPUs are very expensive on the cloud. We predict Akash’s GPUs to be anywhere from 20 to 50 percent lower than what the market offers — that’s an enormous cost savings for highly compute-intensive applications.” — 00:05:45

“As we all know, post-Merge, when it moves from proof of work to proof of stake, there’s going to be an enormous amount of GPU capacity that will be unutilized… the fact that we will be able to leverage GPUs on Akash very soon, it’s a great way for ETH miners to earn additional income on their existing capital.” — 00:05:02

“Us solving the GPU problem, especially by sourcing the underutilized GPUs of the Ethereum merge, is going to be a very important unique value proposition Akash can provide to the world.” — 00:32:09

“After that, we’ll be able to go to more niche use cases with computation heavy workloads — when we have GPUs — things like machine learning, things like heavy games that require a lot of GPUs, where we have an in-market advantage just from a cost perspective… when we offer GPUs, we are going with an amazing value proposition, which is cost saving instantly.” — 00:51:22

“In the next phase we’re going to be focused on data-intensive applications because it’s extremely expensive to run GPUs in the cloud… and especially by sourcing the underutilized GPUs of the Ethereum merge is going to be a very important unique value proposition Akash can provide to the world.” — 00:02:48

“That’s that Ethereum network in itself attracted, right, so 19 billion dollars of GPUs available means we should be able to be in a good position to leverage those GPUs onto Akash Network, thereby reducing the GPU cost on the cloud significantly. So we can’t really wait for the ETH merge to happen, and we’ll have the GPU capability by then.” — 00:11:28

“Phase three is to provide features that are unheard of in web2, like high performance computing with GPUs. We’ll be able to attract machine learning and really high performance computing 20 to 30 [percent] cheaper than the cloud, and that’s a big deal.” — 00:47:46

“GPUs are going to be a paradigm shift along with managed services — that’s going to take [Akash] to the next level.” — 00:54:54

“The problem we’re solving is improving efficiency of global computation, because our thesis is this enormous amount of unused compute out there… conservatively, over 85 percent of world’s compute is not used.” — 00:43:28

“There was a demo, a prototype, that kind of had functionality… it demonstrated that we can run a GPU workload… I can’t really give you a realistic timeline when this is going to come out, because whatever timeline I give you is going to be dependent on upstream Kubernetes to fix their stuff, but we’re making a lot of progress.” — 00:00:44

“Akash as it stands today is the world’s first open market for unused capacity… where the benefit is really the first most visible benefit is the cost — I mean, Akash costs about 95% cheaper than your traditional Amazon, and the consumers of this compute interact with Akash in a fully non-custodial and fully permissionless way.” — 00:03:41

“GPUs are coming within the next three months, we’re very aggressively targeting.” — 00:12:05

“The potential outcome of a globally distributed GPU cluster is just cost reduction to a massive level, and repurposing existing hardware instead of trying to go build new hardware. With the supply chain shortages… if we don’t look inwards, we don’t try to reuse what we have… I think it’s going to be a failing endeavor.” — 00:06:32

“Our next mainnet’s focus is going to be high performance computing, so that means we’ll be coming out with GPUs… our goal is to launch a prototype or get it out by end of the year.” — 00:05:05

“This is the first instance an open market for IP addresses is ever launched, so I’m super excited for it… and fourth is discovery. Our next is going to be GPU, so we’re getting very, very close to GPUs.” — 00:02:52

“With GPUs we’ll start seeing superiority — that means for the first time you’ll see Akash deliver something the current cloud does not have, which is an open market for GPUs… we’re offering GPUs for 20 [percent] cheaper, and this ubiquitous amount of GPUs, all the way from… proof of work miners, after the merge, moving to proof of stake… having very powerful GPUs.” — 00:43:23

“A critical event that took Amazon to the next level is Netflix… similarly [for] Akash, the story is going to be one of the users… doing incredible things to get a competitive advantage over their competitors that are using the traditional cloud… then you’ll see exponential [growth]. I think we’re gonna see GPUs really deliver that Netflix for Akash.” — 00:45:30

“GPUs have been such an amazing resource that has always been constrained because of machine learning and AI… I think this text-to-image generation is going to unlock a new era of art and creativity, and this just needs a lot of compute. We as a society do not have enough GPUs to serve the needs of even immediate adoption for AI. So I’m super bullish on Akash supporting GPUs… giving a world where we have AI that’s much [more] affordable, much cheaper.” — 00:05:49

“There are about 19 billion dollars of GPUs that have been purchased and deployed for eth mining that’s just sitting idle… we are gearing up to have GPU support as soon as possible. I’m super excited to be able to unlock this GPU supply and offer it to the world at a significantly lower cost.” — 00:10:09

“The complexity of AI machines with all these amazing new applications is doubling every four months — that means the need for GPUs is doubling every four months… it’s impossible to even get an Nvidia A100 GPU chip right now. They’re selling double or maybe even triple the retail… $23,000 to $30,000, if you can get them.” — 00:09:26

“A protocol like Stable Diffusion, or a model — it costs about six hundred thousand dollars to get a basic training… with Akash we expect that to be, I don’t know, four hundred thousand dollars, 400 to 500,000 — still a lot of money saving. So HPC makes a lot of sense for us… there’s just enormous amounts of GPUs just sitting there idle.” — 00:34:57

“The need for GPUs is like doubling every four months and we are ready. About 19 billion dollars worth of GPUs are sitting dormant because of eth merge — that all needs to see the market, and Akash is ready to get into this. Don’t confuse GPUs for GPU mining — these GPUs are used for something useful, which is essentially AI.” — 00:58:56

“The complexity and the needs for compute is doubling every four months, especially generative AI… and there’s just not enough GPUs to support that growing. The only way to get these GPUs is to go inward… look at where the GPUs are deployed and try to get access to them, because people that have deployed, purchased GPUs are not willing to sell.” — 00:39:30

“This [is a] massive underserved market that’s struggling. I talked [to] people from big cloud providers to small ones… everybody is struggling to get GPUs. GPUs are still 5x price from their market… and Akash is the only way that can tap into unused GPU capacity… these are general purpose GPUs that allow you to run TensorFlow or machine learning applications — the first time an open platform can do that.” — 00:05:49

“We have enormous amount of GPUs that’s being unlocked from Ethereum mining, just enormous — about 19 billion dollars worth of GPUs. So Akash is very well positioned to capture that market and bring that demand back.” — 00:39:06

“Our biggest demand driver is going to be artificial intelligence, AI, because the complexity and the need for GPUs is doubling every four months and there’s just not enough GPUs around.” — 00:38:21

“So far Akash is just really trying to create a parity game… can we give capabilities of the current cloud provider… With GPUs we’ll start seeing superiority. That means for the first time you’ll see Akash deliver something the current cloud does not have, which is an open market for GPUs… offering GPUs for [way] cheaper and the ubiquitous amount of GPUs, all the way from miners… after the merge.” — 00:07:54

“A critical event that took Amazon to the next level is Netflix… that one story put Amazon on a different planet. Similarly, Akash — the story is going to be one of the users on Akash… doing incredible things to get a competitive advantage over their competitors that are using the traditional cloud, and those stories are going to get Akash… exponential. I think we’re gonna see GPUs really deliver that Netflix for Akash.” — 00:10:04

“There’s a tweet from a16z Andreessen Horowitz’s partner Martin Casado about the repatriation of the data center… there’s one case where he talks about a company saving up to a 40x, sorry, of the cloud bill by building a GPU kit in-house. It still doesn’t solve the underutilized problem — if that company is not using their hardware 100 [percent] of the time, they’re still losing money.” — 00:02:10

“There are about, as far as I estimate in our own interactions, about 80,000 GPUs that were purchased for about 19 billion dollars to mine Ethereum. 19 billion dollars worth of GPUs are sitting idle post-merge… and Akash with GPU capability will be targeting those GPUs.” — 00:20:10

“We’re looking at next three months… we’re expecting next three months, but then all our progress is going to be fully public.” — 00:23:02

“It takes about six weeks from Nvidia to get you A100 chips. You can’t even get them on Amazon anymore because Amazon exclusively reserves those chips for high-paying customers. So if you’re a startup company, you want to do AI, it’s impossible to get that on Amazon. So we’re starting to see the system break down… the only way to reverse that is to take the control back from these cloud companies.” — 00:19:28

“As we’re sort of like going into GPUs in the next six months for Akash, I think it’ll be amazing how we can further actually maybe even provide security to Bitcoin using some of these high-end chips we’ll be tapping into.” — 00:36:01

“GPUs are next. So if you have any GPU-style, proof-of-work-style workloads, I think that’s going to be supported as well very, very soon. So you’ll have a complete node hosting solution.” — 00:29:37

“Today we have something called Nvidia Tesla A100s… right now to get Nvidia A100 it takes about five weeks wait time — there is just not enough chips… I think Amazon has about 20,000 of them, Facebook has 10,000, Stability AI has four thousand… if you go to Amazon you can get A100s — unless you pay them a lot of money, they’re not going to give it to you.” — 00:33:56

“Very few companies are going to control chips, are going to control one of the most important innovations that we have, AI, and that’s not okay. And the demand is just increasing by 10 folds every six months. So to serve that demand you need to decentralize… and share these resources — there’s no reason Facebook can’t share their resources with Stability AI and vice versa when they’re not using that.” — 00:34:39

“It’s going to get even worse because complexity of AIs I believe is doubling every six months… all that means is more complex AI coming in and using a lot of these chips.” — 00:25:53

“We know at least we are working with the top, I would say, two of the five data centers that… have kind of GPUs from Ethereum, and we’re working with a few operating systems as well, and Akash at launch will have at least 20,000 GPUs.” — 00:41:35

“Even the CEO of TSMC in their latest quarterly results came out and said this is going to increase the chip cost by five times. So all our chips are going to increase in prices, especially GPUs.” — 00:27:19

“Oh, it’s going to get worse… there is… two main factors driving the prices [of] GPUs, especially high-end ones… it takes about six weeks to get A100 right now… and the cost about 20 grand per chip.” — 00:24:27

“So far we were able to account up to 900,000 GPUs that could potentially come onto Akash that were all left from the Ethereum merge… in this whole year, in 2023, all the supply will be online, and we’ll have Akash AI as a beautiful product to deliver this value to AI [devs], and AI is going to explode in terms of usage.” — 00:48:10

“The provider incentives will be used to incentivize high-end GPUs like A100s, H100s. We’re anticipating from our estimations that an A100 will cost you somewhere around 50 cents — compare that to a traditional cloud offering which is about four dollars an hour… If you ask any AI dev and offer them an A100 for 50 cents, they will kidnap you.” — 00:20:49

“GPUs are the fuel for AI… the need for GPUs is doubling every 3.5 months as these LLM models get more advanced… GPT-3.5 is 175 billion parameters and GPT-4 is going to be a trillion parameters — that means you need that much of compute.” — 00:21:35

“We’re seeing quite a lot of AI applications even though… GPUs [are] like a few months away — people are just so excited about it… we anticipate AI is going to be a lot more prominent [a] deployment for Akash just from [the] cost factor… [LLMs] are really good at running on high-end GPUs like Nvidia A100s… on Amazon they cost about four dollars per hour; on Akash [they’ll] actually cost more like a dollar an hour.” — 00:20:39

“One of the things that we will be launching the mainnet [with] is H100s… it’s impossible to get them from anyone, and we are able to secure them because people bought H100s and are holding onto them and they have nowhere to sell them except on Akash. So a decentralized system is actually helping them get liquidity on their dead asset… Akash will be the only place where machine learning devs get H100s.” — 00:38:01

“Turns out GPUs are the new oil… if you ask anybody doing any decent amount of training or inference or fine tuning, you’ll notice GPUs are very hard to get… Elon Musk purchased about 10,000 of them a few weeks ago and it’s going to be the thing, right? These, this is the new gold… So we think the future of the way you… the only way you’ll be able to access these GPUs” — 00:03:39

“With H100 we end up being the only cloud provider to have H100… it’s no longer just an alternative cloud that’s cheap, it’s now the cloud that can get [you] resources that you can[‘t] get on the traditional cloud, and that’s going to be the future.” — 01:13:46

“Companies today that are doing machine learning cannot get these GPUs, the high-end GPUs we call A100s and H100s, on the cloud unless you have a special deal with someone higher up in one of these cloud providers… they selectively choose which company should be successful. OpenAI got a big leg up because of this reason and that’s not fair.” — 00:13:17

“Soon I think a real vision is machine learning companies should own their own GPUs, because that’s ultimately where the sovereignty stands… when they’re not using it [they can provide it to Akash].” — 00:28:28

“Nvidia is a software company, not a hardware company — that’s my contrarian take. Nvidia’s innovation is really software, the CUDA SDK that they came out [with] 10 years ago. That’s why they’re really good with AI.” — 01:04:28

“Supply is not catching up to demand, not even close… the demand is exponentially improving, so there’s no indicator that indicates the cost and supply is [improving] over the next five years. And even if they do, Akash is going to be the secondary market, so it doesn’t matter.” — 00:35:29

“I predict GPUs are the new oil. Whoever has GPUs [wins] the AI race. Ultimately OpenAI is better than the rest of the LLMs, arguably, because it has access to GPUs… models are open source, the data for the most part is open.” — 00:31:58

“Ultimately AI is only as good as the supply of GPUs it has, right? GPUs [are] the blood for AI. Whoever controls GPUs control[s] the AI.” — 00:01:26

“The current centralized infrastructure is crumbling [under] server AI [demand]… on the other hand you have the Ethereum merge that basically unleashed 19 billion dollars worth of GPUs that are idle right now.” — 00:03:38

“With the AI wave, it’s going to be, the need for compute is going to go excessively. Imagine I can do 10 times — that means I need 10 times more compute as a human being.” — 00:38:42

“We’re coming from a position of strength — we’re only ones with A100, nobody else has… we’re not coming from a position of ‘hey, we are cheaper’… we have something no one else has, no one before it has.” — 00:25:03

“The market is going to go [9x] in the next 10 years… it’s estimated to, right under 40 billion dollars, it’s estimated to be 400 billion dollars… in nine years.” — 00:50:52

“Chips are the new oil, trust me — whoever has the chips has the power.” — 00:49:29

“It’s gonna cost five times more. The cost of GPUs are going to go high while the demand is going to go high, so it’s going to be crazy for the next few years with this geopolitical scenario.” — 00:48:45

“The chips are going to be scarce and the demand is exponentially gaining, and I think it’s prime for a secondary market to exist.” — 00:52:16

“The kind of demand for AI, considering the complexity of AI models, is doubling every six months. Along with that the demand for these chips is also proportional, and it’s easy to see in the next five to ten years where this market is going — I mean, it’ll be way over trillions of dollars.” — 00:13:43

“The simplest way to understand is it’s an open market for compute, so think of it like Airbnb for data centers… there is an enormous amount of underutilized compute capacity all around the world.” — 00:07:14

“We’re transitioning Akash from like a cheap cloud to the access cloud, just like how we transitioned with Airbnb from a $20 bed-and-breakfast option to now the only place to get a French village… it is about getting access to the resources that you are denied in other markets.” — 00:25:09

“You think about where AI is going: the spend in next five years is expected to be 1.5 trillion dollars — a trillion with a T… and Akash’s market cap is what, 200 million? So if you look at room for growth, there is quite a lot. And again, none of this is financial advice.” — 00:58:31

“A secondary market is very important now to solve the supply chain problem, and that’s what Akash is solving. Akash is the world’s first open secondary market for advanced, high-density GPUs like A100s and H100s. And that’s just a start — the user comes to the chips but stays for everything else.” — 00:35:16

“These chips are the oil — the GPUs are the oil for AI… for the next two to five years we’re not going to be able to solve supply chain problems… right now it takes about two years to get the advanced chips from Nvidia, we’re talking about H100s.” — 00:33:09

“Akash can do 15,000 models, it’s not only one model. So for a comparison, Akash is the most capable GPU cloud out there — there’s nothing that comes close.” — 01:54:27

“I don’t see any restrictions of AMD using Akash, but AMD really I think has at least two more years by the time they can make anything competitive to Nvidia at this point.” — 02:08:14

“Overclock Labs ordered the first set of boxes… so we’ll be able to put the best AMD chips, when they come out, on Akash Network way faster than in any other cloud.” — 02:08:56

“What Nvidia is choosing to do is distribute the chips, fragment the market, so that they have more control over the customer. So the chips are right now distributed to companies like CoreWeave, Lambdas… about 20 of these smaller cloud providers that have direct access to Nvidia.” — 01:48:37

“Sam Altman, the OpenAI CEO, said their biggest limitation to scale the company is access to GPUs… OpenAI is on track to make a billion dollars this year, so a billion-dollar company in revenue run rate, with I believe 30 billion dollars in cash, is unable to get GPUs. So you can clearly see there’s a demand for GPUs here.” — 01:58:04

“You think about where AI is going, the spend in the next five years is expected to be $1.5 trillion, a trillion with a T, right? That’s $1 trillion is $1,000 billion. And I thought the market cap was, what, $200 million? So if you look at room for growth, there is quite a lot.” — 00:58:50

“For the next two to five years, we’re not going to be able to solve the supply chain problems. And on the other hand, what’s happening is as companies get serious about AI, they tend to buy these chips as much as possible because they need these chips to run. So these chips are the oil, the GPUs are the oil for AI. And so right now it takes about two years to get the advanced chips from NVIDIA.” — 00:33:02

“You get an apartment for a year and you only go there maybe two months of the year… You can put them on Airbnb. Similarly, Akash is in the Airbnb for high-end [GPUs]. Now you get into these long contracts because you cannot get them on Amazon… and now what do you do with these unused chips? You put them on Akash.” — 00:14:51

“Ultimately, every AI needs chips, right? Ultimately, without chips, AI cannot evolve. Chips are the new oil, right? Whoever has the chips has essentially the control of the world, right? So access to these chips to be decentralized is extremely critical. And that’s what Akash is working on.” — 00:29:03

“Nvidia alone has 80% market share… Nvidia is fragmenting the market by choice, meaning they are giving priority, lower priority to the bigger clouds and higher priority in the smaller cloud [providers]… Core[W]eave has about 35,000 H100s, way more than Google cloud has. So the tides have turned when it comes to Nvidia H100s and A100s.” — 00:12:31

“I saw it [an] interesting stat somewhere that said one percent of global GDP will be used for machine learning very soon. I believe it.” — 00:20:13

“Right now it’s not the most optimal setup to leverage a globally distributed clusters of 4090s, but in two years I bet you is going to be the case — and that’s really where we’re heading.” — 00:41:51

“Now imagine you paid for something for the whole year which you may or may not know if you have use for it. What do you do? You put them in Akash as a secondary market… if you want on-demand chips, will you come [to] Akash? Because Akash is the only place for secondary markets — essentially an unused GPU market. That’s what the whole thesis is all about. So Akash is in a very unique position to take advantage of this supply crunch.” — 00:36:42

“The moment we launched GPUs three weeks ago, all the A100s are sold out.” — 00:40:22

“H100s… [have] the highest demand. In fact they’re so hard to get, it takes about two years if you order now. It’s harder than a Rolex.” — 00:29:28

“Turns out Nvidia deprioritized the big cloud when it comes to giving them chips, because Nvidia considers this big cloud companies as competition… they prioritize… smaller providers like CoreWeave and DataCrunch and Lambda Labs.” — 00:32:22

“I think that’s going to be a trend: I think 80% of our demand is going to be inference and 20% will be fine-tuning and training.” — 00:30:22

“Copilot is one of those products I use on almost a daily basis… I pay about $8 per month for GitHub and it costs about $20 per month per user to produce it. So on unit economics they’re taking a loss… How do you optimize it? You have to go for something that’s a lot less expensive, and the way you do that is by distributing your clusters and taking advantage of something like Akash.” — 00:15:19

“Every time you hit a prompt in ChatGPT, the time it takes to come back requires one H100 per each concurrent user for that time. Now do the math: 100 million users… that’s talking about 100 million H100s. That’s impossible — they don’t exist.” — 00:16:43

“Last check it was north of 90% for one-week time frames, and about 50% utilization cumulative for all resources… Filecoin has less than 1% utilization of their storage network. So now we have an interesting problem where traditionally crypto has been good at oversupply and less demand — here we have the opposite problem: high demand and less supply.” — 00:10:14

“Akash is a secondary market — it is like Airbnb… you’ll get an apartment for a year because there’s no other way to get them, and now what do you do the 11 months of the year when you’re not in the apartment? You lease it out.” — 00:38:17

“I don’t think we will ever have a way to measure heterogeneous computing… you’re never going to have a way to measure diamonds, for example… that’s why Akash doesn’t have a matching engine, it has a bidding engine — it really comes down to the user to select the provider because the variability is incredibly high.” — 00:11:09

“Nvidia consider these cloud [providers] to be competitors… Nvidia is very, very aware and are optimized to remove power from the big guys… they are distributing the chips to smaller providers — be CoreWeave, be Lambda… CoreWeave has about 35,000 H100s, way more than Google ever has.” — 00:36:51

“Introducing them to additional type of GPUs will open up doors to, call it, innovation beyond imagination — like what can you do when you get access to a million GPUs that have significantly lower cost than what you would normally pay to either buy them or lease them elsewhere?” — 00:25:41

“That’s why all the best applications are always launched on Apple… I think that’s going to be a pattern where the latest and the greatest is going to be launched on NVIDIA. But most adoption is going to happen across different GPUs. It has to, because it’s not sustainable the way it is right now.” — 00:59:28

“It’s very important that the supply, the GPUs, the oil for AI is diversified and doesn’t come from a single company. It’s very, very important. NVIDIA has 80% market share, which is crazy… It’s even scarier how the world is like literally dependent or so critical, reliant on a stable Taiwan, right? If there’s a war in Taiwan, it’s going to disrupt so much in terms of AI that is scary.” — 00:57:54

“Now Akash has probably more than almost all the DePIN networks in terms of revenue… doing that without any incentives, and our utilization on GPUs — for high-end GPUs like A100s — [is] about 95%, and overall GPU utilization is somewhere around 50%.” — 00:18:10

“GPUs launched 90 days ago, and they now account for about 36% of all the revenue that has been ever made in the last two and a half years.” — 00:29:43

“For H100s, which is very very high demand, Amazon charges about $12 an hour and Akash charges about $2 an hour — so that’s literally the cost difference.” — 00:42:39

“That’ll create incentive for providers to bid lower on the prices, so that means you’re going to find A100s significantly cheaper on Akash than any other network, decentralized or centralized.” — 00:19:38

“I think 50% is an ideal sweet spot in terms of utilization rate, wherein the suppliers have enough confidence that they will be able to clear the inventory… tenants have enough confidence that they’ll be able to find compute to scale. That’s the sweet spot — we want to keep the network at 50% and scale accordingly.” — 00:17:21

“Akash is four or five times larger in terms of usage when compared to the Render.” — 00:23:10

“We have to do everything we can, if you want to keep AI alive, to get more chips in people’s hands. Demand is only increasing… but the chip manufacturing and chip supply chain is nowhere close to catching up with the demand.” — 00:35:24

“A million dollar a day, there you go — that’s the big milestone when I will be satisfied that Akash has left the building. Right now we’re maybe 0.3% there. We want to get first 1% and then 10% and then hopefully 100% very soon.” — 00:59:20

“If an A100 cost about $11,000, if you’re making a thousand bucks a month, which is about $1.40 an hour… a year amortization of the chip is very profitable… compared to real estate, compared to any other asset class. If you’re getting back your money 100% within a year, I think that’s very profitable.” — 00:21:36

“When it comes to growth, Akash is growing at 150% month over month, so obviously I’m telling you there is an incredible need for GPUs and compute out there.” — 00:14:29

“If you don’t have $100 million in your bank account and not funded by Amazon, Google or Microsoft, one of the big players, it’s impossible to get high-density compute. That’s just the reality.” — 00:16:39

“Similarly, Akash will have pools. A pool one could be A100 with [NVMe] storage, with PCI card, with X amount of memory, X amount of bandwidth, and certain characteristics of a provider that are preset, and a set of tokens can distribute to this pool… in the early days when there are a few participants, they’re going to get significantly outsized rewards.” — 00:53:39

“Some of the predictions, Gartner, I believe, made a prediction that about 5% of global GDP will be spent on AI hardware. And what is that hardware? GPUs, right? And GPUs are very hard to build… 80% of GPUs used for AI is built by NVIDIA.” — 00:52:43

“About 150 GPUs… but these are high-end GPUs, A100s, high density GPUs, and we have about 600 GPUs coming online every month starting this month… most of them are H100s, A100s and A6000s, and Akash will [be] the only place to get H100s for $2 an hour — which anyone that [has used] H100s knows what that number means — and A100 for 79 cents an hour… H100s starting next week.” — 00:14:09

“GPUs are very hard to get on demand. Why? Because big cloud providers control GPU leasing. For example, on Amazon the only way to get advanced GPUs is to know somebody that knows somebody and spend millions of dollars to even get access to the chips. So if you’re [an] AI startup, there’s no way in hell you’ll be able to get started without having millions of dollars of funding.” — 00:01:33

“There are free GPUs everywhere… a lot of crypto miners that bought a ton of GPUs during the last cycle that are sitting idle… Foundry Digital, which is owned by DCG, is sitting on massive amounts of A100s and 6000s… There are also companies that are raising money with GPUs as collateral, so GPUs are the new spice.” — 00:04:25

“GPUs are very hard to get, and according to Elon Musk, harder than drugs. But there are GPUs available in the wild, and Akash is effectively bringing a lot of that GPU supply to the market for people that need them. Right now Akash, I bet, is the only platform you can get A100 and H100s…” — 00:02:12

“We have 3x the amount of daily average workloads from last quarter — right now about 2,000 active leases, that means 2,000 different applications running at the same time… A lot of experimentation is going to happen over the next — maybe we got till 2026 before the supply chain goes away. So we got two years; whoever has the biggest war chest is going to win, that’s the reality.” — 00:50:05

“Right now Akash is the only place to get on-demand H100s or A100s, which is, honestly, the spice of life. If you think about, GPUs are going to be probably the most important chipsets for humanity for the… foreseeable future.” — 00:08:45

“There is no machine learning without GPUs. GPUs is the new spice, so whoever has the spice can do AI — and we are the spice dealers.” — 00:03:03

“Spice was the most valuable commodity before oil — think about it, the whole East India Company, the largest corporation in the world, went after spice in the 1600s… spice was the most valuable commodity before oil, and now it’s back.” — 00:09:31

“They will not give it to you if you’re hyperscaler. They’ll only give it to you if you purchase smaller units, because they want more customers, they want more diversified customer base… they’re deprioritizing clouds.” — 00:06:29

“Akash should always be the first place to get the most advanced compute, so we’re going to get the Blackwell processors very soon, or GH200s very soon… we’re also looking at ASICs. I want Akash to be the first place to get advanced compute before you can go to the cloud.” — 00:32:07

“In 2018 in our white paper we talked about primary use case being machine learning. We anticipated the need for GPUs… the cost of GPUs and cost of compute was only going to go up for machine learning because the data sets are just getting larger and larger… we specified the primary use case for Akash is machine learning — it just took seven years to realize that.” — 00:09:03

“Akash has more GPUs available than Amazon right now, availability-wise. I mean they probably have more GPUs on count, but not available.” — 00:32:51

“We anticipated the need for GPUs because the data was incredible… the cost of GPUs and cost of compute was only going to go up for machine learning because the data sets are just getting larger and larger. We specified the primary use case for Akash is machine learning — it just took seven years to realize that.” — 00:07:15

“We know scaling supply is much easier with incentives than scaling demand… it’s like literally going to a store and seeing racks empty or racks full. I’d rather have a store that has racks full and less customers.” — 04:06:58

“The price points are so attractive — you can get sub-$2 H100s on demand, which is unheard of… Akash still remains to be the only general-purpose platform that can offer price points at these levels.” — 00:16:32

“People come to Akash to get access to a resource that they can’t get access to. I call it the spice — spice must flow. So fortunately we’re in a position where we don’t need to incentivize further to get adoption.” — 04:00:29

“The price points are so attractive — you can get sub-$2 H100s on demand, which is unheard of… an H100 on Akash costs about $1.46; the same thing would cost about $12 on Amazon, if you can get it… In order to get H100s you need to get into these long contracts. No one’s going to give a researcher [who] wants two H100s on demand.” — 00:16:30

“In 2018, in our white paper, we talked about the primary use case being machine learning… we anticipated the need for GPUs because the data was incredible, it was very transparent — the cost of GPUs and cost of compute was only going to go up for machine learning, because the data sets are just getting larger and larger… we specified the primary use case for Akash as machine learning. It just took seven years to realize that.” — 00:06:31

“Akash will start offering H100s at around $2, which is the cheapest ever you can find. For comparison, Amazon offers H100, if you’re lucky, for $12 an hour. Akash is doing the same exact chip for $2 an hour.” — 00:25:22

“There’s clear demand for AI, there’s clear demand for chips, and demand is actually gaining. I think we’ve gotten predicted, for what it’s worth, that 5% of global GDP will [be] spent to compute for AI… by 2030. That’s pretty aggressive.” — 00:35:35

“From a product standpoint, AI is not even close to being embedded in our everyday products because it’s not fast enough, because there’s not enough chips… imagine ChatGPT part of Siri… imagine the amount of compute you need when you have to power a Siri [on] billions of phones.” — 00:42:56

“If you have to put a prediction, I think supply chain will ease out in next three to five years. I don’t think it’s going to be forever compute shortage.” — 00:37:05

“There’s no real challenge to the most valuable company in the world… I don’t know if there’s going to be another contender that’s going to immediately threaten Nvidia’s position in the next five years… Nvidia is going to have monopoly, as sad as it looks.” — 00:40:41

“We’re still in very early stages, so I would say the next two months revenue is going to be at least triple or quadruple, based on what we’re seeing in terms of supply. We’re supply-constrained now, more than demand-constrained.” — 00:23:57

“There’s obvious crazy demand for these chips, but there [is] also just supply, and turns out the only problem is really coordination — and blockchains are phenomenal coordination networks. In doing so we created the world’s first open source cloud.” — 00:02:10

“Akash is the only way you can get on-demand [H]100. There’s nowhere right now that you can get an H100 on demand. You can get H100 if you go long contracts… and it’s very, very expensive. But if you want to get H100 for $2 an hour, or $1.59, you can get [on] Akash an hour instantly.” — 00:24:28

“We actually talked about machine learning in the White Paper. There was a whole page on it. We talked about how machine learning is only getting expensive… it was very obvious to us, like, that GPUs being the core for AI will be, and the demand for AI as it’s going up, GPUs will be in short supply.” — 00:16:48

“AI compute is predicted to have 5% of global GDP by 2030, I believe — and I believe that’s the case. You’re going to see supply chain ease out… I think next 5 years we’re going to see the supply chain catch up a bit.” — 00:29:28

“Nvidia is a king obviously and their profit margins are incredibly high — the actual cost of the chip is not as much what Nvidia sells them for… well, it’s a monopoly. Break the monopoly — you’ve got to get better or cheaper chips from competitors like AMD… George Hotz is rewriting the drivers for AMD to make it usable… the first ever Tiny Box is going to be on Akash. So things like that are coming, and that’s going to reduce the cost significantly.” — 00:14:32

“To specifically address a question about the $600 billion hole: if you think about it, Amazon makes about 43% margin on compute on average. So if you remove Amazon, you remove 43% at least.” — 00:12:17

“GPUs are nearly impossible to lease on demand. So today if you were to go to hyperscalers… basically they ran out of all the A100 and H100s that you can give, and if they do, they gatekeep them. So they don’t give it to you just because you are paying on demand — you got to pay one year up front and maybe if you’re lucky you’ll get some.” — 00:05:03

“It was supply before, but supply is actually getting better… supply is getting better for training — H100s — but not so for inference, like A100s or 4090s; it’s getting tighter… but incentivization can go far — if you give money to people, they will supply. That’s not a big of a problem. But it’s really on the demand side… delivering demand in a non-custodial permissionless manner is very challenging… the majority of the machine learning researchers have no idea what crypto is and don’t really care about crypto — they don’t even own Bitcoin.” — 00:14:25

“The way trial wallets work is very simple: you get a trial wallet for every user that comes — that alleviates the need to install a wallet or get some AKT to deploy. So we think that’s going to significantly increase activation rate… In the next couple of months we’re going to see a significant increase in demand for Akash because we can solve this problem. It’s easy to convince anybody to come and look at you when you say you have A100 for 80 cents… but it’s really hard to get a user to use you.” — 00:18:02

“We wrote a paper in 2018 describing what Akash Network would look like, and we actually talked about GPUs in 2018, but nobody really paid attention to the GPU shortage then — but it’s very obvious now, seeing the success of ChatGPT.” — 00:07:18

“There’s enormous amounts of chips actually available right now, essentially for free, in crypto miners… when companies upgrade their chipsets… the older models are available — there’s a whole supply chain that buys up these older models and they just sit idle… sitting in warehouses and not being used.” — 00:05:51

“GPUs today are nearly impossible to lease on demand — on demand in the sense, use for however long you need… if you try and go to Amazon today and try to rent a chip, it’s impossible to get on demand, you have to get into a long contract before they can even talk to you.” — 00:02:53

“Who powers these AI machines? You need something called GPUs. How do you get them? You have to go to this cloud provider. And it’s nearly impossible for a company to get these GPUs because they’re scarce. These cloud providers sell the GPUs to the highest bidder. So they literally dictate today how AI should be trained and who should use AI by choosing who they want to give their GPUs.” — 00:04:35

“There are about 8.1 million data centers in the world… in these data centers, they only use 15% of the time. That means 85% of the time they’re not being used… [Intuit] during tax season, their compute is used 95% of their capacity or higher… Off tax season, they only use 5% or less.” — 00:09:13

“Akash again is the only place to get on-demand H100s… H100 is available for [$1.50], which is unheard of in traditional cloud. Akash is the only place you can get that.” — 00:07:17

“Apples to apples, Akash costs about 80% cheaper than Amazon… With GPUs, in fact, H100 on Akash cost[s] [$1.49] whereas the same resource would cost I think $5 on Amazon.” — 00:15:29

“Apples to apples, Akash costs about 80% cheaper than Amazon… and the way you’re able to get this is because Akash taps into underutilized resources, whereas [on] Amazon you’re actually building new resources and new data centers. With GPUs, in fact, H100 on Akash cost[s] [$1.49] whereas the same resource would cost I think $5 on Amazon.” — 00:15:29

“Most compute on Akash is underutilized compute, not cloud compute… same mechanics, same economics as Airbnb versus Hilton. Hilton, they have to sell you at margin, where Airbnb… they have a room they’re not using, so they’ll take whatever they could get.” — 00:35:17

“You just get cheap H100s… which are the most advanced chips for AI for buck forty-nine, which is the lowest in the industry — there’s nothing cheaper than that.” — 00:28:50

“Amazon Web Services… alone [is] a trillion dollar business for Amazon, and profit margins are over 60%, and that’s not okay. You can’t have [an] enormously profitable layer and you call it a public utility. For example, the cost if you look at decentralized systems like Akash is 80% cheaper than what you would otherwise pay on Amazon, and that’s because there’s compute available everywhere.” — 00:09:32

“We are already seeing a decline in Nvidia’s demand, especially as we start cracking being able to deploy or infer on non-Nvidia chips… With MI300s from AMD you’re able to run Llama 3.1 405B pretty well… there’s a company called Exo Labs which seems very promising — they do cluster inference on even Macs. So as the models are getting open-sourced and as we see a lot of optimizations happening for non-Nvidia chips, I think the dominance of Nvidia is going to gradually go down.” — 00:21:49

“As AI gets importance, I don’t think Nvidia is going to stay dominant player in terms of hardware. I think it’s going to get significantly diversified — and has to, because you have one company… I don’t want to live in a digital feudalism where you have one digital lord giving us what we need. Today getting chips is so hard — even the richest people in the world; Elon Musk is famous for saying getting GPUs is harder than getting drugs, even though that’s a low bar in San Francisco.” — 00:23:16

“GPUs is the bread and butter for Akash now, considering the demand for GPUs is just increasing without the supply chain improving, and Akash is now the only place to get on-demand high-density GPUs anywhere on the cloud — and that doesn’t seem to be changing anytime soon.” — 00:01:33

“If you want to incentivize a certain chipset — A100 for example — the way it works in Akash is every time you reach 50% utilization for a certain chipset, the incentives to double the capacity are released… so that way you have a significantly better and more disciplined distribution curve instead of just a broad ‘incentivize everything’ and get thousands of GPUs that no one really uses.” — 00:08:06

“The demand is so high and increasing so fast, the supply is not catching up to any degree of satisfaction. Right now, NVIDIA is backlogged for about 18 months to two years, I believe, for H100s, which is the most widely distributed GPUs from NVIDIA, and even longer for the newer models, the B200s.” — 00:03:53

“For something like Llama 3.1 [40]5B… it takes about 24 H100s to do inference… that can serve about 20 to 30 concurrent connections at high precision… to serve Apple’s users at decent performance, you need a billion H100s. And there are about, I believe, about 700,000 H100s right now.” — 00:05:15

“Assume what the demand is going to be when we onboard 8 billion users, which is going to happen… imagine one H100 per user. And that gets outdated within a year. So then you have newer, bigger models that are more intelligent, that need bigger chips… I haven’t even touched the surface of the kind of demand that I see for H100.” — 00:05:58

“Last year when we did the interview I think the daily spend was $500; today now it’s $5,000 a day — it literally 10x’d over a year… and that’s the kind of growth we want to see at an early stage.” — 00:11:36

“There are lots of underutilized GPUs out there, and Akash has a capability to federate them and offer them at a significantly lower price than what you would otherwise pay for a new GPU. Akash is a used GPU market… it’s eBay for GPUs without a physical delivery.” — 00:08:08

“Nvidia now uses Akash — I don’t think Nvidia uses any other crypto protocol.” — 00:06:42

“If one wants to amortize their chip within two years, I think the price is around $2 an hour, roughly, including the energy cost… so if anything less than $2 an hour, you’re losing money. We have chips for [a] dollar an hour too… that tells you the market is — there’s oversupply of chips now.” — 00:21:57

“As we were deploying a lot of these solutions to different data centers across the world, we realized most compute is underutilized — somewhere north of 85% to 90% of the time, data centers have underutilized compute, unused compute, because you need peak capacity to add peak workload.” — 00:03:02

“We are growing 1700% year-year growth. I think this month alone we grew by 100% from last month in terms of our earnings… If you do that, one order of magnitude growth, for about five years, we’ll reach 1% market share.” — 01:07:35

“After several experiments on Ethereum in 2016, we were convinced that this is possible, so we wrote a paper and published that in 2017 called a decentralized supercloud, and we proposed that there is enormous amounts of underutilized capacity around the world — on an average about 75% of data center capacity that’s active is underutilized, and that’s a conservative estimate; in the reality [it’s] somewhere like 90%.” — 00:02:15

“As of this morning our utilization for GPUs is about 92% — it’s highest of any cloud, decentralized or not.” — 00:19:26

“There is a whole new market for commoditizing, especially H100s. You should be able to trade compute cycles, you should be able to create futures… Certainly I would say like a year, we’re going to have compute market futures.” — 01:05:24

“Nvidia’s dominance is dominant for now, but that’s not long term. Obviously you’re going to see quite a lot of innovation that’s happening in the chip space and the energy and all that — it’s going to challenge Nvidia domination in a very short term.” — 00:14:28

“You can get the same amount of compute that you get from AWS for 80% cheaper, and the reason for that is decentralized systems can reach places where centralized systems cannot… closed-off networks, corporate networks, where there is heavily underutilized compute… about 80% or 85% of the time most data centers are not used, because you need to have peak capacity to address peak workloads.” — 00:11:41

“Andreessen did a study, a paper called the Trillion Dollar Paradox, where they researched S-1 filings of cloud companies, and turns out half of every dollar you spend on one of these cloud services — be it Netflix or Asana — goes to one of the four cloud providers. What this effectively is, is a cloud tax that we all pay but don’t really realize we’re paying, to this group of unelected individuals or entities.” — 00:05:55

“One of the most exciting incentive models for me was, well, if I contribute compute to you, to your model, and if I get an ownership based on my contribution from the model, that means I get inference revenues, because inference is how you make money, and training is how you spend money… if there is some ownership rights that I can get from my contribution, I think we hit the nail on the head, because that solves everything.” — 00:38:15

“這是就是一開始 然後的需求 最好的預測是在數據上的需求 是在Gartner做的 他們說5%的GDP會在數據上的數據上 5% 2030年 那是我們有的需求” (Translation: “This is just the beginning of the demand. The best forecast of demand for data [compute] was done by Gartner — they say 5% of GDP will be [spent] on data [compute] — 5%, by 2030. That’s the demand we have.”)” — 00:00:00

“Over 50% of every single penny we spend on online services goes to one of the three cloud providers. All of us pay a cloud tax without knowing who we’re paying the tax to… [AWS] is alone a trillion dollar business for Amazon, and profit margins are over 60%, and that’s not okay — you can’t have an [ab]normally profitable layer and you call it a public utility… decentralized systems like Akash [are] 80% cheaper than what you would otherwise pay on Amazon.” — 00:05:12

“Our across-the-board utilization right now is 70%, and that was, what, 10% when we began… per GPU right now is about $20 per day, compared to $10 roughly, that was about 12 months ago.” — 00:18:06

“H200s, which are best GPUs to run DeepSeek, are 100% utilized — that tells you that H200s are high demand. H100s were pretty high utilization… so we know that H100s, A100s, H20s, 4090s have high demand.” — 00:15:57

“So far we were able to account up to 900,000 GPUs that could potentially come onto Akash that were all left from the Ethereum merge… In the next year — I think in this whole year, in 2023 — all the supply will be online, and we’ll have Akash AI as a beautiful product to deliver this value to AI devs.” — 00:48:35

“We’re anticipating from our estimations that an A100 will cost you somewhere around 50 cents, compared to a traditional cloud offering which is about $4 an hour… If you ask any AI dev and offer them an A100 for 50 cents, they will kidnap you.” — 00:20:26

“GPUs are the fuel for AI. The need for GPUs is doubling every 3.5 months as these LLM models get more advanced… GPT-3.5 is 175 billion parameters and GPT-4 is going to be a trillion parameters — that means you need that much of compute.” — 00:21:11

“The 10,000 or so enterprise data centers, their utilization rate is somewhere around 15%. So there’s enormous underutilization in these 10,000 professional one-megawatt-capacity data centers.” — 00:07:19

“We’re growing at 10x now every year… we want to get to 10,000 GPUs by end of the year… about 100,000 GPUs by end of next year, about a million GPUs within the next three years.” — 00:44:19

“A big portion of GPUs are actually with the crypto miners, especially the ones that invested quite a lot to mine ETH — post-merge, really, you have nowhere to go. So we have in our inventory about a million GPUs that we have access to in our network alone, in our community, that we potentially want to bring to market.” — 00:25:07

“For someone to really take over the world, they need GPUs. Whoever has the power — GPUs are the oil, right? Like you can have the most advanced tank, but without diesel, there’s no point. You need the fuel for AI. The ones that control that fuel will be in the best position to control the world.” — 00:26:32

“What OpenAI has that other people don’t have is GPUs — that’s where the power comes in. It’s not the model, it’s really the GPUs and the data set. That’s where the power is… OpenAI was able to secure great deals with Nvidia, with cloud providers, to be able to get ahead in the game, but I don’t think they have the advantage anymore.” — 00:22:16

“One Nvidia GB200 NVL72 rack draws about 120 kW… To reach a gigawatt of compute, about 8,300 of these racks — traditionally that would mean a dozen hyperscale campuses. But with Akash, those racks can be anywhere… We’re not bottlenecked by real estate or substation permits. We route around the constraints… Gigawatt power without a gigawatt footprint.” — 00:11:20

“In order to get decent GPUs, you need to have long contracts with cloud companies. They’re going to pick and choose the highest bidders, and it becomes prohibitively expensive. It reminds you of the Middle Ages where you had feudalism… These few lords own the grain, and we all have to stand in line to get the grain because if you don’t, you don’t get the AI.” — 00:40:45

“Our utilization right now is somewhere north of 70%. For comparison, AWS has about 40% utilization… most decentralized networks have super high supply and less utilization — I think Filecoin has 0.5% utilization.” — 00:30:40

“You can purchase GPU, CPU based compute for about 80% cheaper than what you would otherwise pay on the cloud… in many documented cases, publicly, users can deploy on Akash within two minutes.” — 00:00:52

“Since enabling GPUs, the network has scaled roughly from about 150 to about a thousand plus GPUs while sustaining roughly 54 to 55% average utilization across a very heterogeneous fleet — that’s competitive with large clouds and far above typical on-prem. It means few idle watts and more useful gradient steps per kilowatt hour.” — 00:12:32

“There are kids in colleges [who] couldn’t get access to GPUs on Amazon because Amazon would only select the customers based on who pays the highest… one student from Cornell couldn’t get GPUs on Amazon, came to Akash, could get his GPUs, built a company that eventually sold to Nvidia… it’s clear that decentralization is working for people that centralization has failed.” — 00:07:18

“It really became prominent in 23 timeframe when you had GPU crisis. You remember, ChatGPT launched and it was impossible to get GPUs. And Akash was the only one that had GPUs because we were tapping into places where your typical cloud wouldn’t go.” — 00:04:43

“Inference — in order to infer a model, you need one GPU per one prompt. Simply put, every prompt you send… takes 10 seconds to return the answer; that 10 seconds is being computed on one GPU… So if you have 7 billion people in the world, if everybody’s on AI, you can imagine how many GPUs we need.” — 00:13:55

“There is a shortage of GPUs. There’s no amount of GPUs in the world. We looked at Nvidia’s production lines — Nvidia, in ‘23, they made about half a million H100s and we needed about 4 million H100s from the demand… there was such clear, obvious need that basically defined what 2023 and 2024 was for Akash.” — 00:11:47

“I think physical goods are just going to go up… As cringe as it sounds, we’re still in early beginnings, I think, when it comes to, like, resources really — because agents have changed everything.” — 03:11:35

“We predicted in 2020 that there’s going to be a chip shortage, GPU shortage… we predicted two years ago there’s going to be energy shortage… making the bet on Cosmos, right bet; making the bet on AI is right bet; making a bet on energy now is I think it’s going to be the right bet as well.” — 00:20:25

“The winds have shifted from people individually getting individual GPUs to taking large quantities of GPUs… we’re in the process of refactoring our marketplace to support large orders… 50 to 100 nodes, each node having eight H100-type nodes… people are leaning towards long-term contracts versus on-demand… ‘I’m going to pay $2 for H100, I want this price to be locked in for the next two years.’” — 00:16:40

“There is no depreciation in GPUs today… H100s have increased in price. They’re nearly double in price. We’re talking about used H100s… You don’t depreciate, you appreciate cuz the demand outweighs supply.” — 00:24:17

“You can simply put a bunch of 5090s, put it on Akash right now and actually earn a sizable income. Your biggest variable will be bandwidth and your energy costs. That’s what you want to watch out for. Now, if you have access to, let’s say, a solar or renewable, your marginal cost is zero. That’s an incredible opportunity.” — 00:39:59