Carmen Li's Plan to Build a Futures Market for Compute

When we spoke to DRW's Don Wilson last year, he talked about building out a GPU market that might be bigger than oil. Now, a year later, he is working with Carmen Li to do just that. Li is the CEO of two companies — Silicon Data and Compute Exchange (where she works alongside Wilson). The former company is building the index for GPU pricing while the latter is a spot marketplace for GPU procurement. Today's episode — recorded at our live show at City Winery in New York — gets into how Li is building a whole new market for GPUs at her two companies. We talk about the challenge of standardizing compute, GPU price volatility, if used GPUs are like used cars, what goes into constructing a GPU index, and what it means to win the GPU lottery. Read more:Jane Street Plans New Data Center as Computing Power Runs ScarceSpaceX Inks $30 Billion Computing Power Deal With Google Only Bloomberg - Business News, Stock Markets, Finance, Breaking & World News subscribers can get the Odd Lots newsletter in their inbox each week, plus unlimited access to the site and app. Subscribe at bloomberg.com/subscriptions/oddlots Subscribe to the Odd Lots NewsletterJoin the conversation: discord.gg/oddlotsSee omnystudio.com/listener for privacy information.

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00Tracy Alloway?

  There are some market stories where you want every detail. Joe and I have made quite a few podcasts on that basis. But sometimes you've only got 10 minutes and just want to know what's moving markets fast. That's the Barclays Brief Podcast. Every week, experts from Barclays Markets and Research get you up to speed on what's happening and what to watch next. Concise, focused, and brief. Search Barclays Brief wherever you get your podcasts.

• 0:26Speaker 01

  This is the Bloomberg Tech Minute brought to you by ChatGPT. I'm Carol Muster. Getting hired at SpaceX requires exceptional ability and may include having Elon Musk review a candidate's resume. So writes Bloomberg Business Week's Lauren Grush. She reports that the company culture is defined by Musk and job candidates must understand that a commitment to SpaceX means adhering to this culture and exhibiting loyalty to Elon Musk. To get hired, candidates must demonstrate their achievements from past jobs and hands-on experience matters more to recruiters than advanced degrees, with the company looking for people with raw, innate intelligence. A spokesperson for SpaceX didn't respond to a request for comment on its recruiting

• 1:08Speaker 01

  and hiring practices, but according to recruiters, advisors, and those who've interviewed for jobs there, it's the space exploration dream that draws people to the company, and Musk invokes that dream regularly to remind workers why they're there. Prospective hires are told that the job will require stamina, including in job ads that candidates, quote, "must be willing to work extended hours and weekends as needed." One major perk for all that hard work? Employees' access to equity, either the right to buy shares or payment in the form of restricted stock units that vest over time. That's the Bloomberg Tech Minute brought to you by ChatGPT. Get started at chatgpt.com.

• 1:52Speaker 01

  Bloomberg Audio Studios. Podcasts, radio, news.

• 2:08Tracy Alloway?

  Hello and welcome to another episode of the Odd Lots Podcast. I'm Tracy Alloway.

• 2:11Joe Weisenthal?

  And I'm Joe Wazenthal.

• 2:13Tracy Alloway?

  So, Joe, we're still continuing our series recorded from the live show in New York. We had a bunch of great conversations. A couple of them were building off of discussions that we had had previously. And one of those discussions was in Chicago at another live show about six or seven months ago, back in October. We spoke with Don Wilson of DRW about the trading environment, but also about his new venture.

• 2:38Joe Weisenthal?

  Right. And so his new venture is one that actually there's quite a bit of competition in and quite of excitement in. And it's essentially like, okay, GPUs, we know they're very important for the AI boom, et cetera. The question is, can GPU capacity, which is scarce, can it become a tradable commodity such that I can buy futures to lock in my price of access to compute power? Could I sell, resell those futures? Will there be speculators speculating on the up or down price of like an H100, running an H100 Nvidia chip for an hour? This is a big question. We know there's a lot of interest in the actual compute, but whether there's interest in compute futures that's tradable instruments is a very TBD.

• 3:22Tracy Alloway?

  Yeah. And the analogy that everyone always uses is compute is the new oil, right? So why can't it have, you know, a market structure that looks somewhat like the oil market? And there are challenges. Fungibility is a big one. Like, one chip might not necessarily be equal to another chip.

• 3:38Joe Weisenthal?

  Or one chip, the same chip at one data center might not be equal to the same chip at a different data center.

• 3:43Tracy Alloway?

  Exactly. And so even if you're not interested in AI, what I say here is like, the market structure questions and the idea of building an entirely new market is really fascinating to me. And I think others will find it interesting, too. And we really do have the perfect guest. We're speaking with Carmen Lee. She is the CEO of Compute Exchange and Silicon Data. These are the two companies that Wilson is invested in. And they've already announced that they're doing futures with the CME, so really the perfect person to speak to. So take a listen. Last October, we spoke with Don Wilson of DRW fame and he was talking to us about his new project, which was basically building out this compute exchange.

• 4:28Tracy Alloway?

  Now we're here with you six months later. You're actually the one leading it. How far are you in this endeavor and remind us, what exactly are you trying to do here?

• 4:37Speaker 03

  Yeah, so thank you for the the question and great audience. Before I do that, actually gonna call back to six months ago in the Dom podcast you did. You asked them a question, what if compute prices keep raising, gonna go up? At that time, September, October, compute prices were going down across our chips. Now see what happened. I think you called it. I think you called, the market called it. So I'm the founder CEO for Silicon Data. So that's the index provider for GPU indices. We recently announced partnership with CME. So we'll be launching GPU future options on CME in couple of months. Pending CFTC approval, obviously. So that's quite exciting.

• 5:22Speaker 03

  We've been working on GPU indices for past two and a half years, starting 2024 April. So it's been a while. And we launched first world's first GPU indices at Bloomberg Terminal in 2025.

• 5:35Joe Weisenthal?

  Yeah.

• 5:35Speaker 03

  A year later, we launched the partnership with CME. So it's quite exciting. Separately, I heard you mention compute exchange before. So thank you for doing that, Joel. I'm the CEO for Compute Exchange, which is spot marketplace for GPU procurement. So we do reserve contracts, forward contracts, as well as refurbished contracts.

• 5:53Joe Weisenthal?

  Let's talk about the variety of options that we have to financialize compute and so forth. So this, I mean, this came up in our conversation, the first conversation we had with, we had with Ian Dunning. Who is the type of buyer who would want to buy compute on a spot market? Because, right, you talk about typically, we think it's like these multi-year contracts, where some entity enters into a contract with a data center or NeoCloud, whatever. And this, they have this for a while. So who is the mark, who is the buyer or the user of these instruments that might want to buy spot compute or very short-term, short-dated compute futures?

• 6:33Speaker 03

  It's a great question. So the compute market right now for compute change, we have. All our provider mostly are new clouds around the world. Okay. That's one side. Another side is a big variety from AI startup. So even though they are startup, they spend millions of dollars on GPUs. Right. Already. There are enterprises who are traditional businesses, but they are needing a node, two nodes, a few servers here and there for their inferencing. Or I don't know, other deployment needs. They are providers, they are inferencing providers, right? They don't own GPUs, but they provide um open source, open weights, model support for other use cases. So what's a big variety, uh most Amer- North American firms?

• 7:17Speaker 03

  They do a variety of combination contracts. Obviously, on demand give you the most flexibility. You don't pay when you don't use it. However, you're also at the mercy of demand supply curve at any given time. So translate to your price can go from three dollars to six to nine, depends on demand supply curve shifting. So that doesn't help when you can have a predictable margin. And also in terms of scares, you you're not guaranteed for GPU resources for next hour or next month. So you see a lot of people shifting from on demand to reserve even four contracts, right? So four contracts, you basically lock in a deliverables for next whatever month, right?

• 8:00Speaker 03

  Starting in September maybe, or starting in November, February opens, right? So this all comes because of market condition. So computing should cover that physical GPU procurement, also token. So we'll talk about token as well. On flip side, who's gonna use the futures options can be a similar set of people, right? You look at oil market, which we all love double TM brand, right? The people use double TM brand, a lot of them are naturally long oil. So the shells, right, the producers, they need to hedge their revenue volatility by shorting futures or port options. If you're naturally short oil, American line, right? They wanna control their cost volatility, they wanna ob- obviously use future options as well.

• 8:45Speaker 03

  Simple to compute, your NeoCloud or you or anyone have the servers. Ideally, you want to have predictable revenue streams.

• 8:53Joe Weisenthal?

  So the NeoCloud would be the shell in this example?

• 8:55Speaker 03

  Exactly.

• 8:56Joe Weisenthal?

  Okay.

• 8:57Speaker 03

  You have GPUs, right? Or the banks where GPUs on your balance sheet, right? Your long GPUs. Then naturally, you want to make sure your revenue, right? Is stable to a certain degree and then you want to use future to do so. If you are naturally short GPU, which is everybody in this room, unless you tell me you have GPUs, right? Then it depends how much you use. If you want to control your cost volatility, you want to use future to hedge as well.

• 9:23Tracy Alloway?

  Just on the compute exchange side of things. If someone is buying like off the spot market, how do you guarantee? I'm not sure quality is the right word for this, but how do you guarantee they're getting what they expect?

• 9:37Speaker 03

  This is a great question. So I'm gonna flip to slide if you don't mind.

• 9:41Tracy Alloway?

  Yeah, we have visuals for this one too.

• 9:43Speaker 03

  So I usually don't like to use slides, but this time because you mentioned really good question. So we actually call it GPU lottery. So we published a paper early this year at GPGPU conference with Jefferson Lab on GPU performances. We actually, so we can have you create a link and you know to the audience later on. This is A100 by the way. I know we didn't put a tag on, this is A100, 40GB memory bandwidth. We proved this 38% performance variance for the same chip. And then we decompose it into the chip self, intra provider and inter provider. And there's many reason for that, right? And to your point, you can't, you don't know until you get your GPUs.

• 10:25Speaker 03

  We have a plats for GPU, Carfax for GPU, depends how you look at it. So we, in compute exchange, you actually verify the GPU before delivered to you. So basically, you can ask Q for say, hey, I want a 200B 200 nodes. Obviously, we'll give you SPACs back and the commercial back. Same time, independently verify the performances on flops, memory bandwidth, tokens and other information, SLAs and other things. And as a user, you can decide, is price your most important criteria? Maybe it is. Or maybe you're willing to pay a premium for geolocation or the performances that you care more about on latency. Right? We believe give people the option and transparency is the most important thing.

• 11:09Joe Weisenthal?

  Let's stick with the oil analogy for a second. You know, there's a few benchmarks that we all know about. There's Brent, there's WTI, there's others, but those are the two that we talk about. If we transpose this to chips for a second, okay, we say you have an H100 index. We did an episode of the podcast last week, I think, with the CEO of Cerberus, which is another.

• 11:31Speaker 03

  Amazing company, yep.

• 11:32Joe Weisenthal?

  Yeah, but they're a different, another type of chip for inference. Is your assumption that these indices are going to be close enough to the cost such that if you're, okay, I'm running inference maybe on some Cerberus or TPUs or training, whatever, some of these others, that an H100 index will be good enough as a hedging instrument?

• 11:56Speaker 03

  This is the whole goal for me sitting here, actually, right? There's a meaning for every financial products in the functional reason. For commodity, it is for hedging, right? This speculation is great, but it really for people to hedge their relativity, to do risk allocation, to do risk transfer and then asset capital allocation. If we can't do what you said, then we fail at our job, right? So that's why we went all the way back. The way we developed our index model is not simple math. It's not, hey, you have two H100 to simple average, right? Because then you compare apple to oranges. The two H100 can have different CPU, different RAM, different disk, different geolocation, different memory bandwidth. You cannot do simple math.

• 12:41Speaker 03

  What we do is we usually collect six months of historical trading data from over 100 data sources. And we see which factor drives. The price differentiation. So every day, over one hundred fifty thousand traded prices in just our platform. And we normalize the traded prices based on different characteristics of the model itself. And normalize to a base case, and then we do the math of settlement price calculation. Right? So then, this price will be highly correlated, ideally. Yeah. As much as it can to the price you pay at a NeoCloud, for example. However, it won't be the same, just like basis trading. Sure. Right? Like every other commodity is a basis risk. We're helping client calculating the basis risk.

• 13:26Speaker 03

  So you know, hey, you're US East. You may be a bit higher or two than this expectation, a manageable correlation, understanding of the indices.

• 13:36Tracy Alloway?

  You mentioned volatility just then. I mean, the reason people need to hedge is because of volatility. Are you seeing enough of that in GPU prices that like this model makes sense? Because if it's just a steady line up or steady line down, like it's gonna be a kind of boring market.

• 13:53Speaker 03

  So it's interesting. So last year, when GPU price all going down, the big conversation is, why do you need indices for something price will always go down? And this year is why do you want indices when price always go up?

• 14:05Tracy Alloway?

  That's right.

• 14:05Speaker 03

  Literally, this is all the question I get. It's it's pretty fascinating. So when the way we look at volatility, we'll look at daily vol volatility movement, not the price up and down, right? The daily volatility for A one hundred, H one hundred is around twenty to thirty. It's a very healthy commodity volatility range. So it's I don't manage volatility, it just happened to be that volatility that can change. It's all because we normalize it. If you look at each individual chip configuration at different geolocation, the volatility are different. There are some chips with eighty percent volatility, some chips with over hundred. Because normalization of indices, you actually get very healthy twenty to thirty daily vol.

• 14:59Joe Weisenthal?

  There are some market stories where you want every detail. Tracy and I have made quite a few podcasts on that basis, but sometimes you've only got 10 minutes and need to know what's moving markets and why. That's the Barclays Brief Podcast. Every week, experts from Barclays Markets and research get you up to speed on what matters and what to watch next. About the time it takes to grab a coffee. So search Barclays Brief wherever you get your podcasts.

• 15:25Speaker 01

  This is the Bloomberg Tech Minute brought to you by ChatGPT. I'm Carol Muster. Getting hired at SpaceX requires exceptional ability and may include having Elon Musk review a candidate's resume. So writes Bloomberg Business Week's Lauren Grush. She reports that the company culture is defined by Musk and job candidates must understand that a commitment to SpaceX means adhering to this culture and exhibiting loyalty to Elon Musk. To get hired, candidates must demonstrate their achievements from past jobs and hands-on experience matters more to recruiters than advanced degrees, with the company looking for people with raw innate intelligence. A spokesperson for SpaceX didn't respond to a request for comment on its recruiting and hiring practices,

• 16:09Speaker 01

  but according to recruiters, advisors and those who've interviewed for jobs there, it's the space exploration dream that draws people to the company and Musk invokes that dream regularly to remind workers why they're there. Prospective hires are told that the job will require stamina, including in job ads that candidates, quote, "must be willing to work extended hours and weekends as needed." One major perk for all that hard work? Employees' access to equity, either the right to buy shares or payment in the form of restricted stock units that vest over time. That's the Bloomberg Tech Minute brought to you by ChatGPT. Get started at ChatGPT.com.

• 16:48Joe Weisenthal?

  I'm always fascinated by like, you know, we look at the Bloomberg Terminal, for example, and there's a price on the screen and it's just there. And we sort of take it for granted that like it had to come from somewhere. And maybe some, some commodities have like a, you know, there's an existing exchange and a public price. And then there's also a lot of commodities, just bilateral trades. What is the actual process by which you collect the most recent data? So if you say, okay, an hour of H100 usage costs X, right? Whatever it is right now, how did you assemble that number? How did you gather that information from, say, the inference providers?

• 17:28Speaker 03

  So it is a very, can be lengthy, depends on what data sources. The nature of GPU spawn markets, compute exchange is one of them. And then many, many new clouds, hyper-scale marketplaces all have very different contract size, durations, specs, and their way to manage their data, right? So it's a lot of licensing conversation, negotiation, and also context. Look, myself, I don't know, I was used for Bloomberg data. So I was in data business for a period of time. So everything is pretty intuitive to me. It's very important to get a variety of data sources, especially for compute.

• 18:05Joe Weisenthal?

  Like you call them up, like, so it's like, okay, the price is different on a Friday versus one.

• 18:09Speaker 03

  You zoom them up.

• 18:10Joe Weisenthal?

  Yeah.

• 18:10Speaker 03

  You, well, you first have a conversation, say, hey, I love what you do. You're a green new cloud, can I license your data?

• 18:17Joe Weisenthal?

  Okay.

• 18:17Speaker 03

  And usually your, your feedback is, what is in for me?

• 18:20Joe Weisenthal?

  Yeah.

• 18:20Speaker 03

  Right? And I will tell our commercials. And then your concern could be, hey, you know, if I give you all my data, I give away all my secrets. And I will go through traditional licensing agreement. What can I disclose, what I want from you, what I do not want from you? What's the pipeline look like? Are you right, you use a straight bucket job? Are you writing my API to yours? Are you writing to mine? It's a lot of conversation. It's actually pretty standard conversation. And right now we have 8 million pricing points globally around 200 data sources. It's pretty much BAU. A lot of people will say, karma always bring up, can I have your data? Because it's always my ending.

• 18:56Tracy Alloway?

  You know, we were talking about GPU indices and... You're not the only one doing GPU price indices, for sure.

• 19:03Speaker 03

  Not anymore.

• 19:04Tracy Alloway?

  Yeah, not anymore. But when you look at some of the other ones, like sometimes they show different numbers or even different longer term trends. What accounts for the discrepancy there? What are you doing differently or what are they doing differently, I guess?

• 19:18Speaker 03

  So I can't comment out of it was mythology because I actually don't know. Different data, raw data, different mythology will eventually drive for different prices. So the way I would look at this is, you know, it's always smart for anyone to look at multiple data sources and then figure out what is the actual decision you have to make. Which data source do you trust? The market always vote. Um, once the things start trading, the market always gravitates with things actually help them hedge, right? If you're easily manipulatable, if you are not data source, people acting, actions you hedge, that was the point aside from speculation, right?

• 20:00Speaker 03

  So, you know, I love to say, I mean, I also strongly believe we're the best. But again, I will let the market decide, which will happen very soon.

• 20:07Joe Weisenthal?

  So of course, like, yes, there's the economic rationale for the existence of a hedging instrument. And we can understand that someone who is an entity that from time that needs compute, they're short implicitly short GPUs, they want to hedge, et cetera. But the liquid markets also really do need speculators and they need people betting on price. What are you seeing right now in terms of traders or institutions, et cetera, who economically can take both sides of the trade? And how active is this getting where there's just a compute trading desk that is separate from their like economic needs?

• 20:43Speaker 03

  This conversation has been going on for a very long time with various banks, various market participants, speculators, they are all very excited. So some banks obviously have those both sides of the trade, right? So they can cross off some positions internally, that's great. Um, always some they have to use leverage external products. That's where we come in. The way I encourage them to do is I selfishly, I want them to start trading desk on compute. The more people Trade, the better for me, right, selfishly. But same time, it's important for people to understand GPU trading. It's not like you can't just move someone from trade oil or electricity with no background context, jump into GPU compute futures. There's a lot of context where number one GPU, it is not homogeneous product.

• 21:27Speaker 03

  Number two, you have to understand the use cases of A100, H100. Right now, they are not that correlated. Is that right? Maybe that's not right. I don't know. There are use cases which they are pretty separated. But maybe their use cases, they can be transferred. And also, there's software layer to this, right? So right now, you can argue certain use cases, some large models cannot be deployed and the legacy chips. But doesn't mean six months later, you cannot do so. As the software layer compression, model compression gets better, optimization gets better, things can change. So really understand not just the hardware configuration, this local supply demand curve for the server self, also the software layer. That's kind of critical, right? That's really change the supply demand curve and all the way to the user behavior.

• 22:13Speaker 03

  So it's all, it's gonna take some time, so we have engaged with a lot of participants. Yeah. Make sure they have the right setup.

• 22:20Tracy Alloway?

  I have what is possibly a dumb question, but the compute futures, how are those actually settled? Because I have like images in my mind of taking physical delivery of like, maybe one of those big.

• 22:31Speaker 03

  I'll give you a wafer. I'll give you several wafer, that'll be fun. So for the CME futures, will be financially settled? Just like the traditional oil settlement prices goes? For contracts, well, obviously, we do for right now at compute exchange. But we're always open to do, you know, physically deliver futures, especially given we do have Silicon Mark, which is GPU benchmarking. So imagine in the future, you can do, hey, I want a 20 grade A B200 this configuration, this shape of servers in US East. And then at the end, we'll get that. Or usually API calls, so you don't get physical. It's not as cool as physically give you a wafer. But you get API calls.

• 23:11Tracy Alloway?

  One can dream.

• 23:12Joe Weisenthal?

  How do you literally trade it? As in like, let's say, there's probably some very bright people in the room, now within institution, when it's all listed and everything, does it need to go through like a futures broker? Is it like a, could it be like a prediction market? Or you just go to the website? Like, what is the actual? How does someone actually get in this? Setting aside whether they're sophisticated enough and whether they know what they're doing. A lot of people trade who have no idea what they're doing. Yeah, setting all this aside, yes, you know, only trade what you know. But like, what is it through a prime broker? Like, how will people actually be able to participate in this market?

• 23:46Speaker 03

  The beauty of CME is you can do the same thing you're doing now, treating CME products.

• 23:52Joe Weisenthal?

  Okay.

• 23:52Speaker 03

  The same process, same processing margin. That's why you get great margin optimization, right? Everything is BAU. It's no different. We don't have anything right now.

• 24:03Joe Weisenthal?

  So any commodities broker that someone has, they will be able to, on that platform, they will have access to these instruments?

• 24:10Speaker 03

  Exactly right. Yeah, we make it easy for people.

• 24:13Tracy Alloway?

  Would you be upset if a prediction market set up a GPU price contract of some sort? Would that eat into your business?

• 24:20Speaker 03

  Not at all. So we actually worked with Polymarket last year. Someone actually listed my product at Polymarket without my consent. It's always start like that. And then someone told me that. And then we reached out to Polymarket and say, hey, do you want to do something, you know, more real? So we did February settled and April settled a few contracts on Polymarket just to test the water, right? Obviously, we're exclusively with CME right now. But yeah, so I think, obviously, you have to do it right licensing, no middle pillar, right? All the right things, yeah. I, you know, I don't, market can do whatever they want. And then people will choose the best product for them to use.

• 25:00Joe Weisenthal?

  Setting aside the financial instruments for the moment, would people think about AI and they think about the use of GPUs? They mostly still probably in their mind think of like OpenAI, Anthropic and Google, basically. And that's kind of it. But obviously, as you've stated, like the world of entities that serve inference in some form or another is much greater than these three companies that we talked about. Talk to us a little bit more about what the actual world of inference provision looks like outside of the big household AI names.

• 25:38Speaker 03

  So the ones you mentioned, they mostly are closed source models, as we call it, right? But they do have some open source versions, but they're famous for their closed source models. So we actually track 300 open source open weights, closed source models globally from pricing and consumption point of view. It is really interesting if we have actually, you know, we haven't really formally launched LM token indices. You can currently look at Bloomberg and it's on Bloomberg. What's interesting is people are depends, it's all be based on your choices.

• 26:12Joe Weisenthal?

  Hmm.

• 26:13Speaker 03

  Right now, the price actually doubled from our indices from now from December first last year. It's like two dollar twenty one dollar per million token. It's a mixture of input output token prices, average weighted by consumption by buska models.

• 26:27Joe Weisenthal?

  Hmm.

• 26:27Speaker 03

  It's not here, this is GPU unfortunately.

• 26:29Joe Weisenthal?

  Wait, sir, I just, well, since we have this specific chart up right now, what is the Y axis in in this chart showing?

• 26:37Speaker 03

  So you're looking at the the dot per per GPU power rental rate on demand for three chips.

• 26:43Joe Weisenthal?

  Okay.

• 26:44Speaker 03

  The top one, the yellow line is B two hundred new cloud on demand per GPU power. Sorry, it's a mouthful. The line, the yellow line is interesting, right? So every new chip we came out, based on historical data, A one hundred, H one hundred usually came out to be high.

• 26:58Joe Weisenthal?

  Okay.

• 26:58Speaker 03

  And then comes down as more supply, you know, came live.

• 27:03Joe Weisenthal?

  Yeah.

• 27:03Speaker 03

  And then price will come came down and then stabilizes. So that's the trend we have observed for A one hundred and then for H one hundred. So when B two country came out, we polished the data last year at Bloomberg, uh this early this year. The price was high and it came down, which is kind of what I expected, but the slope was less steep than I expected. Hmm, I was like, hmm, that's interesting, that's the slope wasn't as steep. And the and then and then quickly observed the price just came up. And now it's higher than the initial open, whatever you call that, right? Launch prices, that shows you demand supply curve in a different stage than whatever stage we had before. So the A one hundred, the the red line is H one hundred, new cloud on demand per GPU power rate.

• 27:45Speaker 03

  So you can see the price came down last year a little bit, sorry about the scale, so you don't see much, but it came down and came back up quite a bit. Uh I think the last three months came up to like eight percent for the H one hundred. The A one hundred is the older oldest chips among the three, right? They're pretty, you know, uh pretty much a commodity at this point. The price came down, they stabilized, but the price came up about ten, fifteen percent for the past three months. Remember, they're A one hundred, right? They're not the latest and greatest at all. Um, so this also tells you a supply and demand curve shifting.

• 28:19Tracy Alloway?

  Oh, yeah, actually, that reminds me, could you talk to us, because you're doing refurbishment of chips as well, right? Like, which seems like challenging in many ways and kind of reminds me a lot about like sort of Carvana model of compute or something like that.

• 28:36Tracy Alloway?

  How are you actually doing this? Like, how does that business work?

• 28:39Speaker 03

  So, this is cooling two different things. One is for people come to compete change saying that, hey, I wanna, you know, it's you new cloud provider, right? If you get a piece of land, you get an energy allocation, great, congratulations. Then your option is, number one, should I get the latest and greatest? The B three hundred, the GBs, the variable, wait for a few months, or do you want to get refurbished chips? And turn it on maybe sooner, right? Then to you, it's become our I calculation for the most part, right? What's your expected, you know, future revenue generation? What's your residual value calculation? How much you're gonna purchase by, right? It's actually pretty, pretty simple cash flow based, you know, I calculation.

• 29:21Speaker 03

  So, the way we approach residual value and the refurb transaction is, you know, based on I, you know, this is your potential breakeven. Looking at H one hundred, right? Obviously, you're not gonna charge us high speed two hundred, but your cost base is also lower. So you can do the future, you assume a few years of forward contract, you sign in three years. Discount the cash flow back, that's your residual value now, right? So we do that calculation with people, so they understand, hey, what's the value supposed to generate? And then what's the trading, the market prices for refurbished or used GPU? Again, you have the task, you have to make sure things works and it is other nuances to that. But we help with the understanding of the whole residual value. And that's why I hold the whole bubble thing came about, but go ahead.

• 30:03Joe Weisenthal?

  What month was it last year? When like everyone got really obsessed with like the lifespan of chips?

• 30:10Speaker 03

  November, December.

• 30:11Joe Weisenthal?

  Like tweeted something about, he's like, oh, the lifespan, they're like, right? And everyone just spent like three weeks free and then moved on from that conversation, right? Like that was like, what do we know about chip lifespans? Are there misconceptions out there about the how long?

• 30:27Speaker 03

  Yes.

• 30:27Joe Weisenthal?

  These can get be productive?

• 30:29Speaker 03

  I got interviewed a few times, but I don't, I mean, I'm not important, still I'm not important, but back then I even less relevant. I would tell reporters, I was like, look, I don't know what data you're looking at. Um, based on my, I actually have blogs, so my website, which is completely, you can just search for it. Last year, because of that conversation,

• 30:46Joe Weisenthal?

  Yeah.

• 30:47Speaker 03

  Uh, I want you a curve later on. The second year H one hundred residual value, resale value for refurbished chips about eighty five cents on a dollar.

• 30:55Joe Weisenthal?

  Hmm.

• 30:55Speaker 03

  So a year later, you can sell eighty five cents on a dollar. That's pretty good, I would say. The third year is eighty four cents on a dollar. And my, I think my car depreciated way more than that, right? And I drive my car for whatever ten, ten, ten years. So it's, I had the data, but again, I'm not going to argue against narrative, which is.

• 31:11Joe Weisenthal?

  So, but there was a, there's a fairly decent drop from year one to year two.

• 31:16Speaker 03

  That's right.

• 31:16Joe Weisenthal?

  But after that, you see a general level.

• 31:18Speaker 03

  That's November, December analysis. Right now it's a little different. I haven't refreshed the study, but our code is there. If you're my data client, if you run my code, you get a number right away. The another thing I want to point out is L forty's. They're like, you know, the OGs, right? At that time, they used to, there's people still use them. They charge you, hyperscale charge you forty cents per G P U per hour. So, you know, I don't know about two years worth, that number coming from. But I will, I will do that trade every single day. If you sum your two years old at ten cents on a dollar, I will buy it.

• 31:48Tracy Alloway?

  There's a sort of big question looming in the background of a lot of these discussions, which is the B question, I guess, whether or not we're in an A I bubble, right? And you sort of touched on it earlier. You have all this granular data on how people are actually using compute, G P U prices, all of that. What's, what's your take on the big question?

• 32:10Speaker 03

  So as an index provider, I cannot give any forward guidance. Point disclaimer. Nor do I know, right? In fairness, I know what do I know. So the way I look at is, we have defined a bubble, right? So you look at stop bubble, right? And then Nasdaq showed up 200%, it came back down to 84%, whatever back then. That's, it's a bubble, right? The way I look at bubble is, can, is your valuation, can your future cash flow support today's valuation of yours? Right? So then, I'm not talking about like OpenAI and everyone else's valuation. I'm not, obviously, I don't, I don't, I don't understand that, that process. The way I look at GPUs, the machines, it's actually pretty simple.

• 32:53Speaker 03

  Look at future cash flow of your forward contracts, and then you discount it back, can you get your money back for the price you pay? Right? It's actually pretty straightforward in for the machine level.

• 33:05Joe Weisenthal?

  Right.

• 33:06Speaker 03

  But, to your point, right? You can say, hey, what happened if demand dropped? No one gonna use your, whatever things you have. But remember, the forward contract is a signed contract. You, if you have that, you, you kind of know. Obviously, if you have things, the biggest concern is, people have concern overbuilt. If overbuilt, then by theory, then all your prices will calm down because it's oversupply of the market. Right? So then you talk about supply-demand equilibrium. How do we know about the future demand of GPUs, right? I don't know that. Everyone's guess is better than mine, probably. The way I look at it is not, it's not that easy to bring any GPU online. Right?

• 33:49Speaker 03

  You can, you, you hear all those say, big data center bill, $25 billion invested. But they don't translate to immediate GPU availability. Right? You need servers, which you have to be waitlisted, if you buy brand new stuff, colocation, you need optic fiber. So it's a lot of, unfortunately, star has to be aligned.

• 34:10Tracy Alloway?

  But people can default on contracts, right? So even if you have a long-term contract signed like that could not work out. Could you envision like credit default swaps or something in the compute market?

• 34:23Speaker 03

  So that happens in every other markets, right? Every market, if you do OTC trade, you have the risk, some manual default you, doesn't matter who they are, right? So there's various, there's a lot of mechanism to hedge that. The things you cannot hedge is GPU cost, right? The price you entered. So that's something exactly what CME Futures is for. You can have the transparency, the liquidity, and then the easiness of trading in and out and hedge your position.

• 34:49Joe Weisenthal?

  Carmen Lee, thank you so much for joining us.

• 34:52Speaker 03

  Yeah, thank you. This is great.

• 35:07Tracy Alloway?

  That was our conversation with Carmen Lee of Compute Exchange and Silicon Data, recorded live at our New York show. I'm Tracy Alloway. You can follow me at Tracy Alloway.

• 35:16Joe Weisenthal?

  And I'm Joe Wazenthal. You can follow me at TheStalwart. Follow our guest Carmen Lee at Carmen Lee. Follow our producers Carmen Rodriguez at Carmen Armin, Dashal Bennett at Dashbot, Kale Brooks at Kale Brooks, and Kevin Lozano at Kevin Lloyd Lozano. And for more Odd Lots content, go to Bloomberg.com/oddlots, where we have a daily newsletter and all of our episodes. And you can chat about all these topics 24/7 in our Discord, discord.gg/oddlots.

• 35:43Tracy Alloway?

  And if you enjoy Odd Lots, if you like it when we do these live shows and talk about the birth of a new market, then please leave us a positive review on your favorite podcast platform. And remember, if you are a Bloomberg subscriber, you can listen to all of our episodes absolutely ad-free. All you need to do is find the Bloomberg channel on Apple Podcasts and follow the instructions there. Thanks for listening.

• 36:34Tracy Alloway?

  Congratulations, you've built a dashboard. But now comes the follow-up question, and the follow-up to the follow-up, and the dashboard for that. Meet Hex. Give everyone the power to ask their own questions, explore data, and build with AI, all in natural language. Because every answer should spark another question, limited only by your curiosity, not your data team's bandwidth. Go deeper than dashboards. Visit hex.ai.

• 37:05Speaker 04

  This podcast is brought to you by Navy Federal Credit Union. Their Flagship Premier Card gives you four times points on all travel. Earn on flights, hotels, ride shares, and more. Plus, the Flagship Premier Card doesn't require you to use any booking portals and has no rewards limits. It's all the travel, none of the baggage. Visit navyfederal.org/flagshippremier to find out more. Navy Federal is insured by NCUA. Terms and conditions apply for rewards.

• 37:33Speaker 05

  A business gift should do more than check a box. It should reflect your brand and show someone they're appreciated, recognized, and truly seen. 4imprint offers thousands of high-quality customizable products, like premium apparel, drinkware, tech, and more, making it easy to create a gift that feels meaningful and on-brand. And with 4imprint's expert support and their 360-degree guarantee, you can be 4imprint certain your order will arrive exactly as intended. Explore gifting with confidence at 4imprint.com. 4imprint, for certain.