00:41:31 · Vincent Weisser, John Coogan, Jordi Hays
[00:41:31] S3: Amazing. Yeah. So we've announced that we've raised a 130,000,000 and 1,000,000,000 with Prime Analytics Wow. To build the open super intelligence stack. [00:41:43] S1: It's fantastic. Congratulations. Yeah. It [00:41:48] S0: it What I what we were talking about before the show Yeah. Is is the the the the ratio between dollars raised to run rate. Yeah. Absolutely incredible. [00:41:59] S1: Yeah. Right? Much have you raised before this round? [00:42:03] S3: Just 20,000,000 total. So I think we've actually got to this run rate on, like, less than 20,000,000 in spend. That's crazy. [00:42:09] S0: Which is north of which is north of a 100. [00:42:14] S3: Yes. That's amazing. Yeah. We we doubled the pricing since we actually closed the fundraise Yeah. To a 100. So basically, yeah, on track to grow much more from here on. [00:42:25] S1: Okay. So reset me on the latest and greatest in the actual product offering. What customers are coming to you? Is it I have used a frontier model to, you know, figure out that AI can in fact do some agentic workload within my organization? I wanna cut cost and you're gonna RL an open source model for me, fine tune it, and then am I paying you for tokens? Am I are you getting me GPUs? What is the full suite of products that you offer customers these days? [00:42:55] S3: Yes. So basically, it is kind of like the full stack [00:42:58] S1: to do
00:42:58 · John Coogan, Vincent Weisser
[00:42:58] S1: to do [00:42:59] S3: training, deployment, and continue improvement of models. I think customers basically come to us for anything ranging from compute to inference to the full RL and post training stack to to train models. And I I would say it's like, with like customers coming from different buckets and categories, a lot of them are like the AI natives who are basically starting out, and and oftentimes also like being multi model. Right? Like, they they use like the frontier models. They use open models Mhmm. For different use cases. Like, good examples like RAM, for example, as one customer more in this like AI native bucket, a lot of the big AI native startups. But then we've also increasingly been able to track a lot of the new AI labs, like a lot of the Neo labs. So a lot of those are using our full stack from like large scale clusters to our pre training and post training and RL stack. But then also increasingly, like traditional enterprises and even sovereigns coming to us who want to have their own end to end stack that they can run like, on prem. So how I see this is, to some extent, it's like, there's just increasingly, I think, like a demand for having your own, like, sovereign AI stack. Right? It's like your own open end to end stack, but I think this is almost complementing, in many cases, the other, like, frontier, like, models and and stacks. [00:44:18] S1: And you're using Sovereign in not just the geopolitical sense, but also if you're a Fortune 500 company, you want to own your stack from start to finish, potentially.
00:44:28 · Vincent Weisser
[00:44:28] S3: Exactly. Yeah. I think that there was a lot of recent talk about this. Obviously, like folks like Satya and Alex Carr were like, also making making those those points and arguments over the last few weeks. And I think it's increasingly something I think where, like, a lot of, like, enterprises want to basically build this, like, compounding data mode and flywheel, where ultimately, I think the future that we are starting to see is a lot of these companies want to build basically self improving agents. Mhmm. And the way to do this, I think, is generally you you have an RL environment for use case, you scale on it, and then you ultimately deploy that into production, and then continuously improve with the user interacting with that agent. It's almost like the the Tesla autonomy, like levels, but like for knowledge worker agents. So I think like the the ramp example I think was one that people really resonated with for basically them being able to within a week and less than 50 of of training spend Wow. Outcompete a frontier model Yeah. At automating spreadsheets and and finance at a fraction of the cost of of running the cheapest models. Sure. This is like they were able to outperform Opus at a fraction of the cost of Haiku and and like 30% faster, like much more tailored to the use case. So I think in many ways, like, these, like, specialized agentic use cases are a big unlock. And I think in many ways, you still need the god like a big mega model to to do the orchestration or planning. But then I think oftentimes execution, I think, happens can happen with these specialized sub agents. So I think, yeah, those have been extremely, I think, like, fruitful and and and useful for a lot of the companies adopting and and training their own agents. Then I think the other big trend is just like an explosion in in neuro entrants that are, like, taking model training seriously. Obviously, hundreds of Neolabs really got stood up in the last probably twelve to twenty four months. So we started powering a lot of those, like, with our full stack, including with also large compute clusters. So, like, NVIDIA also joined us as part of the round, and we've been like, we've since day one basically operated large clusters. And and this is in in large part why we need so much money. Right? It's like to basically be standing up more and more GPUs.
00:46:42 · Vincent Weisser, John Coogan
Like right now, we run like 15,000 GPUs and scaling to like over 30,000. Okay. And so so this is, like, a lot a lot of the capital needs even. [00:46:50] S1: And walk me through when you're pressing the huge cluster button. Is that because there's broad demand for specific models that you need to inference at scale, and so there's just a lot of inference demand? Or is it that for the level of training that you're doing, you need to be cluster scale to even kick off the training run? [00:47:12] S3: Yeah. So basically, it's like for for all the needs, like, from training to inference, you need obviously compute. In a lot of ways, like, our customers, like, are doing a lot of these, like, large scale training runs with us, like, leveraging our tools, like, including our compute. Right? So it's like pre training, fundamental and parameter models, like, for what we did with the RC, like Q4 last year, or really large also post training runs and RL runs. Right? So it's like increasingly basically, we abstract all of that info away. So basically, people can just, like, hit a button Mhmm. To do a post training run without having to, like, log into GPU cluster and and managing the bare metal GPUs. So we basically abstract a lot of this, like, GPU complexity away for but but yeah. Like, for that, we basically also ramping up our own clusters to basically just, like, serve hundreds and and thousands of customers, like, in parallel on those.
00:47:12 · Vincent Weisser, John Coogan
[00:47:12] S3: Yeah. So basically, it's like for for all the needs, like, from training to inference, you need obviously compute. In a lot of ways, like, our customers, like, are doing a lot of these, like, large scale training runs with us, like, leveraging our tools, like, including our compute. Right? So it's like pre training, fundamental and parameter models, like, for what we did with the RC, like Q4 last year, or really large also post training runs and RL runs. Right? So it's like increasingly basically, we abstract all of that info away. So basically, people can just, like, hit a button Mhmm. To do a post training run without having to, like, log into GPU cluster and and managing the bare metal GPUs. So we basically abstract a lot of this, like, GPU complexity away for but but yeah. Like, for that, we basically also ramping up our own clusters to basically just, like, serve hundreds and and thousands of customers, like, in parallel on those. [00:48:04] S1: Yeah. So what you're saying is, like, there might be a situation where I have a I have a 50 k budget like you gave with the ramp example. I have a 50 k budget to do some post training, and and I need to deploy that 50 k into, like, thousands of GPUs, but only for a short amount of time. And so Exactly. The budget's actually going to be relatively small even though if I were to buy all of that, it would be it's way more than 50 k, millions, millions of dollars. So you sort of absorb some of that cost, orchestrate it, abstract it away, and then you can deliver for the customer. Just a slice of it.
00:48:04 · John Coogan, Vincent Weisser
[00:48:04] S1: Yeah. So what you're saying is, like, there might be a situation where I have a I have a 50 k budget like you gave with the ramp example. I have a 50 k budget to do some post training, and and I need to deploy that 50 k into, like, thousands of GPUs, but only for a short amount of time. And so Exactly. The budget's actually going to be relatively small even though if I were to buy all of that, it would be it's way more than 50 k, millions, millions of dollars. So you sort of absorb some of that cost, orchestrate it, abstract it away, and then you can deliver for the customer. Just a slice of it. [00:48:38] S3: So so basically, I think there's this, like, also big unlock, like, also on the inference side, right, where customers don't need to, like, front load hundreds of millions and and CapEx spend for large data center commitments. [00:48:49] S1: Yep.