Intelligence on the Edge: Liquid AI's Ramin Hasani on the Search for Device-Native Foundation Models

Hello, welcome back to The Cognitive Revolution, and happy 4th of July to everyone in the United States. Today, my guest is Ramin Hassani, CEO of Liquid AI, a company founded by MIT researchers that's developing device-native foundation models. I'll say upfront that just before recording, I encouraged Ramin to go deep into the weeds on the technical details of Liquid AI's work, And as you'll hear, he did a truly excellent job demonstrating a mix of technical sophistication, differentiated vision, and a contagious passion that, in my humble opinion, makes this episode an instant classic. We start with an overview of the team's research into tiny, biologically inspired, differential equation-based neural networks that Ramin and team developed at MIT, and which inspired them to start the company. Some of the capabilities they demonstrated, such as parking a car with a control module that consisted of just 12 liquid neurons, still sound a bit like science fiction today. And while those systems haven't scaled up to today's capabilities frontier, the company has maintained the liquid philosophy. Which today means taking a neutral, empirical approach to designing and optimizing neural networks to perform under all sorts of exotic constraints. Including most commonly the need to run on edge devices with limited memory and processing power. Considering the fact that the global smartphone and laptop market is roughly $800 billion per year, a number which the global AI data center buildout is only now surpassing, the demand for inference threatens to price much of the world out of the frontier model market, and that so many enterprises and individuals value privacy and the ability to control their own information, this is an absolutely massive market opportunity unto itself. And Liquid has serious proof points, including holding the number 5 spot on the Hugging Face United States downloads leaderboard, plus notable partnerships with companies such as Shopify and Mercedes-Benz. Anyone who doubts can do a quick download and demo of Liquid's Apollo app, which shows in my experience that even a 1 billion parameter model, which combines a small number of attention layers, a very simple gated learned convolution— while admittedly far from the frontier— can run fast enough on an iPhone to be a real option for basic use cases, such as privately searching through and classifying one's own local documents. Perhaps most interesting is the network architecture search process that Liqid uses to develop networks for particular use cases and runtime environments. Having found that proxy metrics too often lead the process astray, they now evaluate models on real downstream tasks on the actual target hardware that their customers intend to use. Romain shares a lot more detail on their findings, but in short, while attention-based architectures continue to generalize better than any known alternative and therefore continue to dominate the frontier, the more specific your use case and the more limited the compute resources you have available, the more likely their search process is to to land on an exotic architecture. This, for now, is where architectures like Mamba and other subquadratic innovations really shine. Toward the end, Ramin teases a platform that Liquid will soon be introducing to allow customers to fine-tune small models for their own use cases on a self-serve basis. For multiple reasons, including the potential to ease demand for frontier models and improve access to AI globally, I, for one, will be very excited to see that come online. And so without further ado, I hope you enjoy this high-energy look at how Liquid AI is squeezing as much intelligence as possible out of any given computational resource with co-founder and CEO Ramin Hassani. The Cognitive Revolution is brought to you by Mercury, the fintech that more than 300,000 ambitious companies and individuals trust to run their finances. I've wired AI into nearly every corner of my life. My email, my messages, my calendar. I even gave Mercury virtual cards to my agents with low limits and category and merchant restrictions for their autonomous use. But still, my AI's access to my financial data has remained limited. With a normal bank, I might export a bunch of statements and have my assistant process them for me. But for real-time, up-to-date information, And certainly for taking any action, trying to get your agent to use the bank via the browser is just too hard, too slow, and too error-prone to be worth it. And that's why Mercury's new conversational interface, Command, is such a big deal. It's built directly into Mercury, which means you get natural language access to your finances without exposing anything outside of your bank account. No exports, no spreadsheets, no pasting your transactions into third-party tools. I really think a lot of people are going to prefer it this way, and it can already help you take actions too, with everything bound by the permissions and approval policies that you've already set up in your account. I am genuinely impressed to see this level of AI integration in banking in 2026, and so I invite you to join me in the future. Visit mercury.com to learn more and apply online in minutes. Mercury is a fintech company, not an FDIC-insured bank. Banking services provided through Choice Financial Group and Column NA, members FDIC. Thank you to Mercury for supporting The Cognitive Revolution.

And now on with the show. Ramin Hassani, co-founder and CEO at Liquid AI. Welcome to The Cognitive Revolution.

Thanks for having me.

I'm excited for this conversation. I've been following Liquid AI from afar for a number of years. And fascinated by some of the architectures that you guys have developed going back to your time at MIT together. Also fascinated by the trajectory that I perceive the company to have taken over the last couple years as it's become more customer-focused and commercial entity. So maybe for starters, how would you tell the kind of broad story of Liquid AI? Leading up to what you're doing today and what the company's mission is today?

Yeah, absolutely. So Liquid AI, 3 and a half years ago, we spun out of MIT CSAIL, building on a technology that we have actually been working on it like 10 years before, basically like a decade before when we started the whole company. Our objective function at MIT has always been maximizing the amount of intelligence that we can into smaller format of algorithms. So efficiency has been like the cornerstone of our research. And, uh, we have been working on, you know, like specifically thinking about robotics and systems that were like coming into the real world, you know. Like the idea of liquid neural networks, that was kind of discovered also by my PhD thesis. And together with my co-founders, the 4 co-founders that I have, we have been, we've been researching ideas about how can we build machine learning solutions that can go on robots and doesn't have like millions of billions of parameters so that we can actually host them directly on, let's say, CPUs, NPUs, or let's say like, you know, like smaller GPUs that are mounted on top of like physical systems while delivering the reliability of much larger kind of instances of artificial intelligence systems. Essentially what we, what we try to do, we try to, you know, like build alternative algorithms in order to get creative in the algorithmic space to see how How can we build machine learning systems that can generalize beyond the data that they have seen? Because when you go in the real world, the distribution shifts becomes like a real thing. Imagine you deploy a robot in an open world, kind of like, let's say, an autonomous car, a flying drone, a fixed-wing vehicle. All of these systems are going to go and very, very rapidly you're going to get out of distribution. You have to actually build systems that are really comfortable around being able to get out of distribution. You know, humans are extremely good at it and humans do that extremely at an extreme— I mean, natural learning systems, animals also like the same way, right? So they actually follow a really nice trajectory out of distribution as well, you know? So that means like the learned concepts can generalize to data that you have not seen before, right? So our idea was that like, can we build systems that can have that kind of properties, you know, better out-of-distribution generalization? This has been not just our goal. I think an entire artificial intelligence field has been working towards this thing, and especially in the robotics and closed-loop environment, like closed-loop video environment where you have an agent acting in an environment, you want to have that kind of properties. So naturally, the place that we started looking into was brains. We started looking into animal brains, and we started looking into how— one thing that I really liked, I wanted to look at it from a first principle kind of approach, like how neural neurons exchange information with each other. You know, we started looking into brain of worms, you know, small animals. And, and why worms, this, this specific worm, C. elegans? We started working on the brain of the worm. The reason behind it is that it was the only animal that in 2015, when I started this type of research as part of my PhD with my co-founder Matthias Lechner, we, um, this was the only animal that we knew the entire nervous system as a whole. And this animal exhibits massive amount of kind of sensory reactive kind of behavior, like amazing levels of control with 300 cells in its nervous system, you know. And that was fascinating for us because this is much smaller than any neural network that performs control at that time, like on, let's say, like autonomous systems, you know. And this worm, it can do better, you know, dexterous movements and stuff like better than the best robotic systems that we actually had in the world, you know. So We thought that, okay, so let's start understanding how neurons exchange information in the brain of the worm. From there on, let's start building nervous system more complex and complex kind of neural circuits so that we can get to the stage where we can, let's say, like build the next animal, build the next, you know, like basically follow the path of evolution of nervous systems, you know, and see how we can rise and evolve as part of this thing. So, There are equations that describe the neuronal dynamics of, let's say, like 2 neurons in the brain of C. elegans. Because the worm is very small, the neurons do not spike. They are graded kind of neurons, like they behave like in an electrotonic way, you know. So they're very similar to artificial neural networks that we have because they're also very differentiable, right? Because you don't have spikes in the activity of neurons, they're very differentiable. That's why, like, it was even nicer for us and more attractive for us because we could apply learning theory to these type of neural networks once constructed. Like, let's say 2 neurons, 4 neurons, 8 neurons, like, let's say 100 neurons next to each other, and then you start, like, training them. You can apply backpropagation, like, as a differential kind of programming on top of a system that is built by these type of inspirations that we got from nature. The type of differential equations that were there were also very well behaved. You know, you can make them as complicated as you can, you know, and And then when they get more complicated, they mimic the biology much more. And, you know, of course, like, from a computational footprint, they become more complex to scale, you know, but it's still like you can sacrifice, you can actually tune how much complexity you want to encode inside a neural architecture. You know, in artificial neural networks, like, we try to abstract away, like, those complex differential equations that describes the neural dynamics between 2 cells And we just show them with a sigmoidal function, gated sigmoidal functions. And now we have like matrix multiplications that are coming, like capturing the impact of like the inputs that are coming to a system with a transformer-based architecture and attention mechanism and all of those things. But these are all simplified kind of computations for us to be able to scale these machine learning solutions. For us, from a neuroscience perspective, it was very interesting for us to explore to see if we actually start making, like, bringing back those differential equation-based computation and a little bit more elaborate form of computation into the behavior of every single neuron and mimic the behavior of, like, how 2 neurons exchange information with each other, maybe we can unlock something, like, greater than, uh, what, what we have seen from an artificial neural networks perspective. So this, I mean, early on the results were fascinating. Like, we saw that with 12 neurons, we actually 12 neurons, you can— you could parallel park autonomously like a car, you know, like a small car. With 19 neurons, you could drive a car. With 30 neurons, you can fly autonomously like navigating kind of a drone, you know. So, and you can get sensory information and process this information with these a little bit more complex and elaborate version of this neural dynamics, you know, which we called liquid neural networks. And liquid for adaptability. I called it liquid liquid time constant neural networks. Like, that was the LTC kind of paper that we, we got out. And we coined the name liquid for the fact that the dynamics of these systems are staying flexible even after training, you know. So the models would be able to react to new type of inputs that they receive and even learn during backpropagation how to be adaptive towards the inputs that you're receiving, you know. So it actually encoded A little bit more degrees of freedom on flexibility of the learning dynamics of a neural network compared to artificial neural networks, compared to other systems that you've seen. That form of dynamics actually allowed us to scale this technology into not just, you know, not just in robotics, but also like applying it to predictive AI in financial services, in medical domain, and in many, many different places that we applied in pre— like, let's say audio modeling, you know, multimodal video understanding, you know, like we applied this technology over and over.

Mm-hmm.

And we saw promise that this technology is actually very, very amazing. Like, you know, every neuron is much more complex than a normal kind of neural network, but at the same time, like, you don't need that many points of computation in order to actually get to the results that you want. We also saw that the out-of-distribution performance of these models are extremely good. So, and they were suited for robotics applications. So, I mean, we had interactions with the United States Air Force. I mean, there was at that time, Boeing was actually hosting my postdoc at MIT. And there we showed that you can actually fly even jets, like, with this type of neural networks with, like, a handful of these neurons, you know, which was very interesting kind of topic to see how far you can push, like, let's say, out-of-distribution generalization with a handful of, like, in a very, very small set of system, you know, a system that is, like, powered by these liquid neural networks as opposed to artificial neural networks. Then, so as I told you that each node in a liquid neural network is very complex, you know, like it's, it's a differential equation. You have to solve that and you have to solve a set of different— the more neurons you have, the more complex is a forward pass and a backward pass through a network. Therefore, we needed, we needed to think about efficiencies. You know, these, these systems are highly nonlinear. You know, like we try not to sacrifice on nonlinearity because adding nonlinearity to learning systems allows you to build more expressive systems. You know, this is a proof that actually I have in my PhD thesis where I showed like nonlinearity actually directly contributes to, especially on the smaller size of models, you know. But then the idea here for us was that like, how can we scale these systems? You know, you have nonlinear systems, but it is extremely difficult to actually scale nonlinear systems. So you have to solve— the reason behind scalability is the computational complexity of the models are hitting like even cubic complexity. You know, it's a it's not even quadratic, you're talking about cubic complexity, which is like too much. And then you have like diff— like you, you usually have like, when you have a set of differential equations, you can roll them out with a numerical solver. You run the numerical solver and you can step by step compute kind of the, the outputs that, that are desired. And then, you know, like the, the more steps you, you want to use in order to actually compute, then you would get more and more accurate kind of results at the end, right? But then the problem becomes scalability again. You know, like, you, you cannot really get to infinite precision with these numerical solvers. One idea that came to us was that what if we just solve the whole system in closed form? You know, like, literally, like, just you have a differential equation system with these liquid neural networks, each of them representing neural dynamics of, let's say, 2 cells that are exchanging information with each other at a certain level of abstraction. And we saw that, okay, Let's take this system and try to solve it in closed form. Turns out the closed-form solution for this type of equation has— hasn't existed since 1907. You know, so 1907, there was a scientist called Louis Lapicque that actually modeled the membrane potential, like how to model mathematically membrane potential kind of in, in, in cells. And that format of equation became like a fundamental of channel modeling, right? like how information propagates through ion channels inside, like inside a cell, and then how neurons, how neurotransmitters are getting propagated to the other one. So 1907, this was Louis Lapicque's kind of membrane potential equation that is like an open differential equation. And then really we've seen like some scientists called Hodgkin and Huxley, they started working on really biological kind of grounding this type of differential equation and adding a little bit more complexity into the model. They developed a neuroscience model of one neuron, how actually a neuron reacts to, let's say, a membrane potential. 1953, they started. 1963, they won a Nobel Prize for the fact that they actually developed a better and a more accurate representation for neuronal dynamics, right? And then from there on, so that if in every textbook that you read, these type of formats of equations, they do not have a known closed form. form solution, you know. And this format, like liquid neural networks, were also part of that type of equations. We, for the first time, we actually solved that. And in— this was like 2022, around 2022, we solved the liquid neural network kind of interaction of neurons with each other in closed form for the first time. This became a Nature Machine Intelligence paper published in November of 2022. This is called the closed-form continuous-time systems, you know, like this is liquid neural networks in closed form. And, uh, the closed form, I mean, it has massive implications. Why? Because now I don't need to use any numerical solvers to actually run a liquid neural network. I can now have not only hundreds of neurons, but now I have— I can have billions of neurons next to each other, and I can actually scale this computation still keeping the nonlinearity as part of this thing, you know. Then, then in, in January of— actually, actually beginning of February of 2023, an article came out of a Quanta magazine about, let's say, me and my co-founder Matthias. It was about a profile about all the implications of having some closed-form solution finally on neuronal dynamics and how important this can be for both machine learning and also for brain science as a whole. And then, so my inbox was full of VCs. Silicon Valley started talking, everybody wanted to throw a term sheet at us to really get started and really scale this technology because this was fundamentally different than a transformer-based architecture, an attention-based architecture. This was grounded in biology, like the type of math that we have. Around the same time, we have alternative models are coming out like state-space models, like, you know, like the— you know, like you've seen like faster iterations of state-space models, like convolutional neural networks that came out like to be like scalable. All of them in a linear form because the scalability of alternative models You have to linearize them and then you can scale them. And we have something here that is also another category. It adds some operators that we learned from biology and from physics as we grew the research. And we got to the point where we can now become very competitive and building models for solving more and more general-purpose tasks. And when I say more and more general-purpose tasks, I'm talking about modeling signals beyond predictive performance. modeling signals for language, for audio, and for vision in a way that humans understand. That's the complex format that humans understand. So the mission of Liquid AI became building efficient general-purpose AI systems at every scale. And we coined the word efficient in our mission, and we care about it so much because I think we are, from a foundation model company's perspective, we are probably the most efficient foundation model company on the planet. I'll tell you why. About this in a minute, but the reason why, you know, like we set the mission into like efficient machine learning, because we saw that how much of the considerations that we had to do in order to really build a closed-form solution, then taking this closed-form solution and scale it to, let's say, a system that can model language, a system that can model audio models, a system that can model video and vision in general, you know, how much considerations had to go into these things so that we can, in a computationally tractable way, scale our neural architectures to the point where we are today. Today, the base of our technology— I'll talk about the core technology of the company later on, but I will tell you today the technology became liquid foundation models. Liquid foundation models, or LLMs, are pretty popular right now. We are ranked number 5, actually, in the US from the number of downloads. from the popularity perspective, we have over 1 million downloads per week on Hugging Face. That's a lot of downloads of these small models that we are building. They're very popular. The top ones from the US side are Google, Meta, Microsoft, and NVIDIA, and the 5th one is LiquidAI, so from the US side. We got ourselves there by consuming about 1,000 GPUs, having 1,000 GPUs in-house. That's from a foundation model perspective, getting to that level of popularity and releasing more than 50 models, instantiation of models that people are using in enterprises. That's kind of the place where we brought Liquid, building on top of inspirations that we got from nature and, and, and the past that I actually like portraying for you.

Hey, we'll continue our interview in a moment after a word from our sponsors. Today's episode is brought to you by Anthropic, makers of Claude and Claude Code. Over the last few months, Claude has helped me build and refine a personal deep context database. that now contains all of my emails, Slack messages, tweets, DMs across platforms, video calls, and podcast transcripts going back a full 5 years. On top of that, we've now layered summary articles describing my relationship with hundreds of contacts, organizations, and ideas. And now that this exists, there's almost nothing that Claude can't help with. For my angel investing, Claude can now draft investment memos in exactly the form that my venture fund requires. Based on the calls I've had and the emails I've exchanged with the founders. And when someone needs a favor, Claude can often do it as well as I can. Recently, a friend reached out to ask if I know anyone who might be a fit for a role that he is currently hiring for. Initially, nobody came to mind, but then I thought to ask Claude, and sure enough, it identified 2 great leads. Claude is the AI for minds that don't stop at good enough. It's the collaborator that actually understands your entire workflow and thinks with you. So for problems worth solving, get started with Claude at claude.ai/tcr. That's claude.ai/tcr. And check out Claude Pro, which includes all of the features mentioned in today's episode. That's claude.ai/tcr.

Brilliant. So many different directions I want to go and follow up there. Maybe for starters, one thing that just on a very fundamental level jumps out at me is when you describe the liquid neural networks, you describe them in terms of the number of neurons, as we're used to hearing about the number of parameters. And something tells me there's a kind of paradigmatic difference there that underlies the difference in description. So yeah, help me develop my intuition a little bit more for— I guess it probably connects to nonlinearity. And this is so fundamental, right? Because there's so much incredible progress that has been made, but on things like really robust out-of-domain generalisation, we've still got a lot of work to do. On things like adversarial robustness, we still have a lot of work to do in the mainline paradigm. So Yeah, tell me, unpack a little bit. Why are we counting neurons versus parameters and what does that tell us?

That's a great question. So I think early on we actually started talking about the number of neurons as the unit of computation, like for in our research, because that was like very interesting. Like you can associate, it's a, you know, like every neuron is a process that can be modeled by a mathematical equation, you know? Now you can allocate basically like in a sigmoidal kind of from gates. So the more functions you have, the more parameters getting added to that single cell. cell, you know. And the cell that now in a liquid neural network, in terms of parameters, you can say multiply by 7, you know, and that would be the number of parameters of an neuron, you know. So that in that differential equation. But that cell itself, because of the equations that it has, it computes more than just, you know, like a forward pass computation. It has like internal feedbacks. It has 3-degree feedback kind of mechanisms inside. All those parameters are actually like one-to-one analogous to any artificial neural network system that you would build in terms of parameters. So a rule of thumb for when I say number of neurons, you multiply it by 7, you would actually get the number of parameters of the system. But this is for early-on version of liquid neural networks. As we started standardizing parallelization schemes for the type of mathematical functions that we want to explore and, let's say, inspirations that we're getting, We just converge to the number of parameters. Of course, today when we are talking about a liquid foundation model that has, let's say, 1 billion parameters, we are literally meaning 1 billion parameters in the sense of any other kind of GPT that is getting built out there. Does that make sense?

Yeah, but let's zoom in maybe a little bit more on that neuron. And I guess I'm wondering a little bit too, what is the fundamental limit, or what are the first bottlenecks that you you hit when you try to scale the original liquid approach. With the sigmoid-type functions that we have today, those are, I think, certainly carefully selected to be easily run on GPUs. I'm guessing that these sort of multiple degrees of freedom inside a single neuron maybe present challenges in terms of executing that process on available hardware? We can get these amazing results from tens of neurons, so the obvious bitter pill question would be, what if we take that exact paradigm and go to millions, billions? But you must be hitting some bottlenecks along the way. What are they?

So the main challenge is turning sequential computation into parallel computation. So when you have degree— like when you start going from a single neuron dynamics to multiple neural dynamics, weight parameters of your system, instead of being scalars or vectors, they become matrices and tensors. So now you're talking about matrix multiplication. You want to turn the scalar computations into tensor computations. You want to be able to— this is how you parallelize let's say sequential computers. The problem is that not all the time, like if you have nonlinear relationships in an equation, you cannot trivially create a one-to-one map between a vectorized computation and a tensorized computation. So those nonlinear kind of nonlinear attributes of liquid neural networks and any other nonlinear recurrent neural network, I mean, recurrence itself adds like some degrees of complexity into the math of the whole equation. I mean, let alone if the recurrence has a nonlinearity on it, it becomes a lot more complex to disentangle vectors or write them with typical linear algebra methods that we know into tensors. You have to be able to turn these computations into tensors to be able to compute in parallel. You compute them once at a time. That's the mathematical— it's not related only to liquid neural nets. Any mathematical operation that you want to tensorize, If it is nonlinear, you're going to have the troubles that you're talking about. That's why the state space models that actually came out, these are all around linear dynamics. We're talking about linear dynamical systems, right? And why those linear dynamical systems are important is just pure fact that we do not have proper ways to scale nonlinear systems. You can apply nonlinearity to linear dynamical systems. For example, in a, let's say, a gated, or let's say, you can add a sigmoidal function after you perform the dynamical system in a linear way. You compute all the matrices and everything in parallel, but then you can do a pointwise application of a nonlinear operation on top of the entire tensor. That's what you can do. But what if the relationship between the parameters themselves are governed by some nonlinearity? So you cannot really use typical linear algebra. You got to always approximate that nonlinear system into a linear system. then you would be able to actually parallelize the systems. Right? Does that make sense? So that's the fundamental bottleneck.

So how much does the closed-form solution address that? And how far has the— with available computing resources, how far has the original Liquid network paradigm been able to scale Fantastic question.

So, one of the properties of liquid neural networks was the fact that they have multiple feedback mechanisms. They're not just one format of feedback. They have 3 layers of feedback between 2 cells, in the synaptic dynamic itself, because we wanted to mimic how it is done in the brains. So, it has multiple degrees of feedback. It has then a nonlinear— What is feedback? Basically nested nonlinearities on top of each other. That nested nonlinearity itself would also add degrees of complexity. Even the closed-form solution version of liquid neural networks, from a dynamics point of view, you can now, instead of 100 neurons, you can have 100,000 neurons, maybe 1 million to 10 million neurons. But the nonlinearities, you still have to compute the models sequentially because of these nested nonlinear relationships that actually happens in the circuit. So even the closed-form kind of solution would be limited in scalability in the sense that you cannot parallelize them very effectively. But there are also like— people are, people are also like working on these topics a lot, like parallelizing, but not just parallelizing, but the speeding up sequential computations. So we're talking about sequential scan, you know, like Scan is an operator that you can run on top of, let's say, doing matrix multiplication. Instead of doing it in cubic or, let's say, quadratic, you can do it in sub-quadratic time. You can actually reduce the complexity of computations even on nonlinear operators without parallelizing them, without linearizing them. There are ongoing research going on. We are doing some of that research ourselves to really find ways to to really sequentially run those nonlinear systems because clearly there is an advantage to have like those form of nonlinearities when you do function approximation. You know, we are talking about also scaling laws becomes also very important here. I believe like scaling laws define architecture. You know, how does it do that? When you're talking about transformers being this revolutionary thing, we are talking about maximum scale. The reason why transformer architecture and attention mechanism is such a brilliant architecture is the fact that it is unstructured. There's no structure. When I'm talking about nested nonlinearities and all this crap that we have, like in liquid neural networks, you don't have that in transformers, right? You have basically just matrix multiplication as the core functional kind of things that represent, and it is unstructured. You can literally multiply Any matrix into any matrix of any size. So the whole idea here is this: the larger neural networks that you make into infinite size, the larger neural networks you make, the more you want them to become less and less structured. And we've seen like the success of Transformers at the trillions of parameters. Now we are talking about tens of trillions of parameters. Like, I mean, the next generation of models that are going to come. We're talking about trillions of parameters, and you can do that with a transformer architecture. As soon as you start adding a little bit of bias in that architecture at scale, things become completely messed up. So we are talking about right now liquid neural networks, like these alternative architectures that we are talking about. They have a scale to hit, you know, like there is a regime of parameters that you can just do better, you know, like let's say up to, let's say, 100 billion parameters. Up to a trillion parameters, you know, like that's kind of the range we are operating right now. In this range, the smaller is the model architecture, the more you want to specialize them for a certain application to solve. And they are actually like mathematically, they're biased to actually solve a certain type of tasks better than any— better than other types of architectures. So I would say biases on algorithms, you know, the more, the more kind of biases you put on there. And what by biases, what I mean is this Adding a lot more nonlinearity, adding like multiple gating levels on top of a neural network, you know, adding like recurrence, you know, like multiple different types of recurrence, adding, let's say, convolutions. Like, as a— I mean, convolution itself, if you just keep the convolutional kind of neural networks, they're also pretty unstructured because you can apply convolutions on any size of neural networks, you know. So that, that, that question becomes like how much bias and what kind of problems you want to solve. So that the bias in the neural architectures becomes kind of a function of, I would say, scale of the neural networks as well as like the use cases you want to solve the most. You see, so the spectrum would be the more you go to the larger size, the more unstructured kind of mathematical operators. Use pure matrix multiplication, use pure convolutions, use pure, let's say, sequential scan, pure, pure operators. You see, it's like you get much more like— and forget about gates, forget about adding fancy control, like forget gates and all those things. And at scale, this is something that we have seen because we have scaled neural networks and we have a really good understanding of that architectural variation at different scales.

Just for calibration, when you talk about the liquid neural networks that have gone up to hundreds of thousands or pushing a million neurons. What kind of hardware does that run on? Is that like a CPU supercomputer?

You can do that on a CPU, like on even simple CPUs or simple GPUs, because they don't have that much of— probably like one of these liquid neural networks would fit on, I don't know, 1 to 25 megabytes of file size, basically. You can literally put them there and then you know, a CPU would be enough, like a Raspberry Pi would be enough, like, to perform computations. We have shown that on a lot of predictive specialized applications of AI that these systems can be, like, very, very powerful and do stuff. And they can be as powerful as the systems— like, these closed-form variants could be as powerful as the open differential equation version of the neural networks in speech synthesis, you know, like, that's one of the applications. In, as I said, predictive sequence, like, imagine you have complex formats of sequences coming in, multivariate kind of sequences that are coming in, and you want to perform some sort of a prediction on top of them, these neural networks are actually pretty good. And then if you're talking about more, let's say, out-of-distribution generalization, to the extent that you are within a bounded period, you cannot have an open-ended continual learning. These are not— liquid neural networks are not continual learning systems. They are more adaptive formats of computations. because of the many gatings, many feedbacks, and many, let's say, input-dependent parameters that they have in their mathematical kind of organization.

What is the definition of continual learning that you're using there that they don't satisfy?

So continual learning would be a system that is continuously receiving new data and retunes itself. So it changes also the parameters of the system. Liquid neural networks are the dynamics of the systems are input dependent. So that means new data comes in, the dynamics of the system would behave to those inputs, but the parameters of the systems, like any other neural networks, is fixed. It's just because the, the format of every neuron or every node in a liquid neural network was like a differential equation, you would have a different dynamics when you're— so dynamics, you know, it's another dimension that you add to number of parameters. It allows you to compress more information, compress more knowledge. That's why with the smaller instances of the models, there's no free lunch, there is no magic here like we are doing also with liquid neural networks. It's just that the axis of dynamics has been something that we added to the neural networks. For what? For being more adaptable. So I'll give you a very tangible example here. Imagine you're driving and all of a sudden it starts raining. So depending on the format of the rain, your autonomous driving system that is actually taking control of the car could react to those driving scenarios. If you have a traditional neural network that hasn't seen that environment, it might actually get biased because that rain that actually hits the camera, it's noise, a certain type of noise or a certain type of adaptation of the input that is coming. But the reality hasn't changed. The confounding variables of the whole environment hasn't changed. So that's why liquid neural networks, because they react to the input completely differently, you know, like they absorb the input, they apply the low-pass filtering on top of the input, you know, they are much more adaptable to those kind of scenarios. That doesn't mean that they change the parameters of the system to be more adaptable because there are 2 axes here. One, changing the parameters of the system. The other one is changing dynamics of the system. I would consider continual learning to be attributed to where you continuously also change the parameters of the system. In liquid neural networks, we don't do that. Gotcha.

Okay, helpful. Okay, so let's fast forward to the present. You guys are now in the market working with customers. One notable conversation I heard with one of your customers was with the CTO of Shopify, on the Latent Space podcast, who had some very good things to say about you and your technology. I was struck by the fact that you've gone neutral in a sense. Now, excuse me, what sort of jumped out at me in terms of your approach is that you've developed an architecture search process where the promise to customers is not that, hey, we developed this one paradigm and old calculus teacher used to say, when all you have is a hammer, everything looks like a nail. So you're explicitly promising to people that even though we came from this lineage of this particular kind of network, we're not just going to blindly apply it to your problem. Instead, we've created this sort of higher abstraction or more meta process for searching through architecture space to find the thing that's going to work best for you. And the 2 notable Details of that are, one, proxy metrics you found don't work that well, just measuring perplexity or whatever. You found you needed to actually go farther and test models on the actual downstream tasks that they're going to be asked to perform. And second, hardware in the loop, actual target hardware in the loop to test the architecture Subject to the very real and physical constraints of the robot, the sensor, the phone, whatever it's going to be running on. Maybe I understand that the LFM model came out of that, but maybe before we even get to the LFM, could you sketch out the range of different types of problems that we're putting into this architecture search? And then maybe also a subject that implies, of course, what kind of hardware are we targeting? And then what kind of different architectures are winning for different kinds of problems under different kinds of constraints?

Absolutely. So there's a system that we developed in-house. We call it Automated Foundation Model Design, AFMD. That's a meta-learning system that puts the hardware in the loop and then tries out many different operators. with an evolution strategy, the criteria is an evolution strategy, optimizing for a couple of things, optimizing for memory consumption on that device, optimizing for latency, optimizing for speed, while no sacrifice on quality. When we talk about quality, perplexity is not the measure. It's actually the application, the downstream applications that we care about. It's not just also public benchmarks. We are talking about 100 different benchmarks. So the problem space becomes, from a meta-learning perspective, becomes like a very, very complex kind of problem. Now, I'll tell you that why did we take this approach to design an architecture? Why? Because we wanted to remove all the human biases early on as we are actually building architectures. One of the things that we realized culturally at companies, this is what I can tell you, even at largest foundation model labs in the US right now, Anthropic and OpenAI, there are a bunch of people that are coming from the science. I call them the Avengers of the architectures or Avengers of the post-training, or let's say pre-training, these groups of people, there's usually a very small set of people that are calling the shots on like, oh, you know what? You're going to tweak this portion of this architecture so that it performs better.

Why?

Because in my personal experiences, it has started working better. If you're really truly thinking, and this is something that is broken in all the foundation model labs, you cannot say that somebody has a fix to this, but now the recursive self-improvement process is actually fixing for that. Because now people are just finally realizing you got to give it to the algorithms. You have to be bitter lessons. You have to— bitter lesson pill. You got to be giving it to a systematic way to actually find out what is the true architecture for the problems that you want to solve. You can build a general-purpose computer. The insights that I shared with you in the format of the scaling laws of neural networks is coming out of our massive exploration.

Yeah.

exploration of the space of architectures. The fact that in the smaller category of models, some biases on the architecture help. In the larger instances of the models, you don't need to bias the systems. You can actually go pure convolutions, pure transformers, and you can do pure, let's say, recurrences that are very simplified. You don't need to add any form of specialized treatment like gating there, gated DeltaNets, and then there's the Mambas and there's the Jambas and there's like— There's a lot of different variations of these architectures that are coming out. What you want to do, though, you want to be completely unbiased. Even us ourselves, day one, when we actually started Liqid, we had one of the inventors of SSMs actually as our founding scientist, Jimmy Smith. He invented S5, like the simple— he basically brought the parallel scan idea to the SSM world. Then we had Stefano Massaroli and Mike Polley, they were actually designing the hyena hierarchy. There was a convolutional path that is going on. All of these are linear systems, within the realm of linear systems. Then we came out of these nonlinear hyper-control theory enabled foundation models, which is liquid neural networks. A lot of big heads in the room and you got to call the shots around. Again, the same problem in all the labs. figure out. We thought that, okay, let's fundamentally solve this problem. Let's really systematically, first principle, let's put all the operators of interest, whatever we think that could be an operator that can give rise to a general-purpose computer, like a system that can— let's say a liquid neural network on its own is a general-purpose computer. You can actually scale it. If you can theoretically, you can get it. So you put all those kind of equations, all those kind of things in a unified kind of theory. You know, like with linear input varying kind of operators, you know, like operators, all, all the, let's say, space of operators can actually be encapsulated inside this linear input varying, or let's say in general liquid input varying kind of operators. Because input dependence is— was something that we have been like talking about this for the last 10 years, you know, like that's, that's extremely important. And we see that right now in transformers is also extremely important to have that input dependence is actually coming naturally with the attention architecture as well, you know, like, but in a very, very unstructured way. It's not as structured as, let's say, like what is done in RNNs, in liquid neural networks, in, you know, like in SSMs, you know, like in all of those things. You don't see these gated variables like to be very structured in transformers, you know. Now we actually have various variants of convolution operators, various variants of of a recurrent operator, you know, and then the attention themselves, you know, there's like, there's group query attention, you know, there was the original transformer, there's like many different things. Then on all some of these variations of dynamical systems, and also various types of convolution, I said convolution before. So, but then what we brought, we brought also the liquid computational blocks, like for example, the gated, the double gated conferring, so gating what, like, a certain type of biases on, onto, like, different dynamical systems or different operators. We added them, and probably the space becomes like something around 50 to 100 different operators. And then you want to build hybrid models so that you can reduce— basically, what's the goal here? The goal here is that maximizing efficiency of computation without loss of accuracy. That's kind of the goal that we set. The objective function of the search space is also this. let's run. Let's have all of our— let's say all of our compute actually thrown at this problem, and let's try to see how the system is going to design this. And we started doing scaling laws on this. We went as small as a neural network as, let's say, 10 million parameters to running the scaling laws up to 72 billion parameter models in these hybrid structures. So we have done a massive— early days of LIQUiD, 2023 and 2024 has been always like proving out on a certain type of processor what's the most efficient type of architecture that can come out. And turned out, as, as, as all of our, uh, you know, when we put all of our biases away, you know, all the gating stuff that we are putting around operators, you know, they got simplified. So, you know, like in Mamba-style architectures, like you have a bunch of gating mechanisms in let's say, gated DeltaNets, you know, gated DeltaNets, you know, you got— you have like a bunch of operators in there, you know. In linear attention, there are like gated variants. Everybody's like tweaking a certain parameter in the network like by hand because in their own experiments actually they observe something. Turns out all of this has to go away if you want to get to the most efficient form of— format of architecture. And it became the double-gated convolution that actually came out of this massive search space like AFMD, like this original kind of system that we designed. And this was one of the candidate architectures that came out that turns out to be very good on general-purpose computers that are called CPUs. On CPUs, because CPUs don't have— they have a special kind of a structure, right? On all the CPUs coming out of AMD, Qualcomm, let's say Intel, everyone that is building processors, ARM, all the ARM processors, We tried to get to the place where test the different operators, what's that generic structure, and let's say computational graph that gives you to the most simplified, no added hand-tuned, there's no hand-tuned features to anywhere. It's just literally coming out of the test, the semantic test that we've done, and this became the de facto architecture Elephant 2 structure that we actually announced. But while exploring these things, like, we've observed, like, many different candidates also popping up, you know, like, there's, like, too many things. And then we try to, like, really, like, figure out, for example, for, let's say, an NPU, a neural processing unit, that is, like, inside an AI PC powered by AMD or powered by Qualcomm or powered by Intel, which of these variants would be, like, a better neural architecture that gives those hardware providers and silicon providers a boost on the amount of efficiencies that they are unlocking, or let's say speed of computation, latency that they get, memory footprint that they get, while having a computational graph that doesn't sacrifice any format of quality. So this was like the whole thing, removing all the human bias with a systematic approach to architectures, even our own biases, you know. And the only thing that actually remained in Elephant 2 on CPU competition is this double-gated As I told you, we have nested computation in liquid neural networks originally. This nested format of computation was something that actually became very interesting to be part of these very simplified neural networks that we built. The fact that we have unstructured 1D convolutions as the layers of choice, 70% to 80% of our networks are structured by these gated convolutions that we have, double-gated convolutions that we have. They're extremely simplified. And they replace attention, they reduce the computational complexity by a lot, they reduce the memory footprint by a lot, they maximize kind of the speed of computation by a lot, and at scale as well. And at the same time, on the quality, I mean, you have seen like some of these models are actually like extremely kind of competitive to the transformer-based alternatives. I think you asked a bunch of other follow-up questions as well, but I would pause here like for any follow-ups here.

Yeah, let's get into use cases in a minute. I'm definitely interested in that. And I also want to talk about the future of hardware and what your work implies for the future of hardware. But I think it would probably be helpful for a lot of people, including myself, although this is something I've studied in some depth, I still would like to grok it better than I do. This concept of gating, I'll give you my rough and ready understanding and then you can improve it and deepen it. I especially noticed this with Mamba when I went down that rabbit hole and became very excited about it. It seems that the kind of trick that is played over and over again with these gating mechanisms is we want the transformation that is done on the data to be input dependent. So it's not enough to learn a transformation, we want to learn a transformation, but then also have some relatively— typically the gate is relatively low dimensional. Sometimes it could just be like a scalar that's applied to that transformation. It could be obviously more complicated than that, but it's some relatively simple mechanism that says for this learned transformation, here's how we're going to modify it given the input currently under consideration. And that seems to be great. It seems to unlock a tremendous amount. Help me understand more anything you think I'm missing there, or help me deepen my intuition for why that is such a powerful and recurring theme in all these different architectures.

That's a liquid structure. That's when I say, like, this input-dependent kind of thing that is actually coming into Mamba. So, right before Mamba actually got out, actually not right before, but a year and a half before Mamba came out, we released a paper called Liquid S4. This paper was— if you just read the abstract of that paper, we are actually for the first time, we are introducing this idea of input-dependent SSMs, basically input-dependent SSMs. The idea there was that let's bring the fundamental, let's say, building block that we we found has a lot to do with representation learning, with capacity of learning, this input dependence element to really bring that format of gating to the neural architectures, to SSMs, and then bringing that to convolutions as well with the other systems that we are designing today, liquid foundation models as well. That format of gating, it actually adds a lot of— just think about it, like, naturally it makes a lot of sense. If the neural networks— it's not about the forward pass of the neural network that it is adaptive, you know, like, because once when you're input dependent, you have, you know, like, input dependence in the learning— from a learning theory perspective, every— all the magic happens in the backward pass, you know. If you think about when you're computing the gradients backwards, that input-dependent kind of operator itself It's going to represent itself. So when you're learning from that data that you're seeing, you're also learning some sort of dynamics, input-dependent dynamics there. So that second axis that I was talking about, it's not just the number of parameters of the neural networks, it's also the dynamics that the neural network actually learns. You have some representation of that dynamics being learned in the backward pass. Now, the complexity of that gate is going to make a huge change. Also, it turns out that gate is extremely important also in language modeling, especially when you have sequence models like RNNs. Let's say classical controls, like continuous-time RNNs, let's say discretized RNNs like LSTMs, and then you have liquid neural networks, again, very nonlinear versions of these linear versions of recurrent neural networks is an SSM. When you add this type of gating, you, you can improve like the language capabilities as well. It seems like for, let's say, discrete sequences, this gating also helps a lot. You know, it's not just for the continuous time sequences but also discrete sequence, because I would consider language to be like a discrete, uh, sequence, right, of information. So you would have like that sequential dynamics that is kind of added by this, by this gating. So it is a very right frame of mind to think about that kind of feature. And for us, like, we think that this input dependence is one of the fundamental discoveries that we have done, like, that we learned also in biology. It happens, you know, like, it comes out of physics. Like, if you just put the math together, you're going to realize, like, oh, what is special about this dynamical system is the fact that it has this nonlinear input dependence. That's kind of a structure. And by biases, I mean the complexity of that gating. Even existence of that gating is a bias that you're adding to your system, you know. Now, is it needed or not at scale? That's a big question mark. Like, you're going to see when neural networks are 100 trillion parameters, is matrix multiplication enough, like, to really, like, perform, like, general-purpose computing, like, get to AGIs of the, of, of the world or not? That's, like, a big question mark. But also, all, like, you need to also put this discussion about architecture and the obsession about architecture in the— in perspective. Another perspective you have to put these things, the learning theories themselves. You know, right now we are talking about different learning schemes that are coming. For example, you can train a model with next token prediction, you can train them in a word modeling kind of context, you know, you can train them in a sequential, in a long-term horizon and long-term history. You know, there's, there's so many ways you can construct also the objective function of a learning algorithm itself, you know, that, that would also contributes a lot to the learning process. Architecture, I'll tell you that we found the purpose of the architecture and the most important application of architectural kind of research has been in efficiency, right? Really getting to efficient formats of computation without loss of quality. That's an extremely important thing because we are talking about the resource allocation problem right now. As the models are becoming bigger and the demand of AI is exponentially going higher, you want to have more efficient versions of the systems to actually run the systems at scale. Otherwise, how can we actually provide access to AI to all? And we are seeing early instances of this thing that the larger labs are wiping out the compute off of the planet. Why? Because they have to train and they have to then host these models for all of us. So efficiency becomes a fundamental property of these model architectures that we are doing the research on. I would say if you want to get into architectures that enable next generation of neural networks— sorry, next generation of intelligent systems in a way that they can become like human brain that actually performs, you know, like performs computation with 20 watts of power, and it's already an AGI that does that between— that's like the massive exploration that goes beyond architecture. It goes into memory research, it goes into learning algorithms kind of feature, it goes into data data feature, you know, it goes into prior research, you know, like looking at it from a statistical learning perspective, that becomes kind of a— and then learning theory itself, like the limitations of learning theory itself, it also gets imposed on today's learning systems because the definition of learning theory itself is actually broke, like, at scale. Like, but what we have right now, like, everything is IID, you know, like we are, we are building averaging machines, you know, you've seen the quality of writing and the sycophancy that actually comes out of AI systems because of the fact that we haven't gotten— maybe to some extent, multi-agents has been becoming the solution and test-time scaling became a solution for the caveats that you're seeing from just learning systems with autoregressive modeling as a pre-training method. But I think there will be innovations needed, not just on architecture, but the whole thing as a whole. You have data, you have algorithms, you have data, you have models, and you have learning algorithms. Altogether, they can design basically the future, that ultimate kind of holy grail, basically, in this space.

Yeah, that's very interesting. So basically, your perspective is that as we get into the recursive self-improvement era, the real advances will come more from new learning paradigms, new objectives, and then those advances will be made efficient through architectural optimization. But the architecture comes after the paradigm question of what exactly are we learning and how are we giving a signal.

It is one component of it. That's what I would say. Looking at it from a data problem, data representation, from, let's say, an axis would be architecture and the other axis would be just the learning algorithms themselves, right? And then we get into that recursive self-improvement, which is kind of the continual learning kind of characteristics of the new system. The fancy name for the continual learning kind of research that has been happening for more than probably 4 decades now. I mean, and that's—

Yeah.

So, and that's how I would actually characterize the whole space. So we cannot just look at architecture in isolation as the fundamental thing that actually changes everything.

You remind me a little bit of Ali Behrouz's The Illusion of Architectures. If we have time, maybe we can touch on nested learning. But let's stay focused on your work for the moment at least. So in LFM2, this is the result of this architecture search. And the kind of surprisingly simple thing that comes back is some reduced but still critical number of attention layers. And then the other layers are— and we've seen this kind of with SSM-attention hybrids as well— but I think the kind of surprising revelation from the result of this search process is The non-attention layers can actually be extremely simple as long as they have some gating. So you've got the gate and then you've got just a real simple convolution that I think I read only considers a very short span of tokens, right? Is it like just—

They're short convolutions.

Yeah. And so if we were going to update the headline, attention is all you need, from however many years ago, now, we would maybe say attention is something that you really do still need, at least at a certain scale, but you also need gating on your other layers. But you don't actually need anything like super crazy fancy sophisticated, the state-space model, all that kind of stuff. It turns out that keeping the gate is actually the part that really drives the most value. And then you can have a really simple mechanism behind the gate. And then you can have 70% of that and 30% of attention. And subject to obviously resource constraints, that ended up being the winning formula. Am I getting anything wrong there?

No, I mean, you're touching on the right things and this is basically, it's like the regime you're operating in, the goal of your system. Are you trying to build superintelligence? Are you trying to build like the most powerful version of the AI system? You need the most unbiased version of an algorithm. Now, attention is an extremely rich, unbiased format of algorithms. Even if the computational complexity of attention is n to the power 2, maybe we really need n to the power 2 to really get to that kind of level, you know. And maybe even we need like more complex architectures. We've always like tried to actually reduce the complexity of architectures for the sheer purpose of the fact that we are resource constrained. As humanity as a whole, we are resource constrained right now. So I would say the discoveries that we have right now, it just shows that there is a gradient on architecture that you can follow as you scale models. The gradient that you're following is the fact that for smaller kind of models and specialized models, you can put as many biases, like these gating mechanisms that you're bringing in, and as many— You can play around with as many operators of interest in your computational graph. It is going to work and it is going to give you some sort of a boost if you're really maximizing for linear time complexity. You want to implement linear attention systems, just the fastest kind of— if the speed is so important and you're actually wanting to even sacrifice a little bit quality, you can bring in linearity and the whole system could be linear. Yeah. You don't even need some of those hybrids. But hybrid just boosts that accuracy because as we see, as I said, the O, basically the complexity, that computational complexity at a certain level, at a certain scale, it is needed for us to really get to those performances that you want to do. The larger the network becomes, the more unstructured you can make it. That's the learning from that whole algorithmic approach that we started designing neural architectures.

Yeah, very interesting. Okay, let's look at the other end of the spectrum then. As you work with customers, what are some interesting examples of when, given resource constraints, given the narrowness of the domain of interest, other kinds of bias are actually winning in the architecture search process?

Great question. So, for example, if you go to, let's say, biology, and you want to model sequential data in biology, we're talking about DNA data in biology. DNA data, from a vocabulary perspective, they're very limited. They're not like language that is maximum amount of vocabulary and stuff. In DNA language, they're very simplified, but the lengths of those sequences you have to actually process are, let's say, for a human being or maybe for a bacteria, I heard, It's somewhere between 1 to 100 billion sequences, like basically elements in the sequence. For that long context kind of things that you want to perform, when you do not have that much of a large vocabulary, you don't need attention. So you can actually run on these kind of data, you can run pure convolutions, pure SSMs, pure liquid neural networks, like system-like recurrent neural networks and parallelized version of these recurrences and linear attention. For extremely long context, you cannot do it any other way. The reason behind it is because context lengths become so large that that quadratic cost of attention just kicks in. So let's say on biological data, you would want to have some sort of structure there. Then there are places like video modeling and stuff. In those kind of places, let's say you might want to have different architectures and even different learning algorithms. You have seen this, success of diffusion, for example, there, right? Like, you've seen, like, diffusion is also, like, a— I would say, like, this format of a prior that you put on a certain architecture. But it's still, like, you— there's debate between, like, diffusion and learning algorithms, or diffusion is actually part of the architecture. You can actually make that connection, and you can make that distinction as well. So you've seen, like, on video and scene understanding, probably elements of diffusion would be needed. You know, like, we— some people still believe that with autoregressive kind of modeling, you could actually get there anyways, but we will see if that stays true. Then when we are talking about audio signal as well, like if you audio alone, for example, if you're talking about audio alone and language is not part of the whole thing, it's just pure like voices, like voice-to-voice, or let's say noise-to-clear kind of signal, you know, signal-to-signal, these are places where recurrent neural networks is still very, very powerful. They're extremely powerful, especially in the low data regime. I would say models that have a lot more biases in them, like in these smaller kind of regimes, when you do not have that much data, this is kind of the place where you can bring a lot of value. And recurrent neural networks and biases in architecture can help you in the low data regime to really fill out that gap with the feedback mechanisms that they have. The more complex architectures, the more you would be able to handle. The closer the architecture is to the dynamics of the datasets that you're trying to solve, the better of a learning system you're building. In fact, in the physics modeling, when you're talking about— there was a class of models like physics-informed neural networks. These are kind of Again, another architecture that we are talking about here. They would be really good for physical kind of simulations, you know. And, uh, yeah, so this is the other side of this spectrum that I'm talking about, like the property of the data, the, the lengths of the, the context lengths, and, you know, like, like all the other considerations that I said would change the architecture by a lot, you know. And then there are— again, you want to scale these to the largest kind of regime, make it again unbiased. Again, a transformer would be able to, at scale, beat this. But at a smaller scale, transformers would not be able to beat any of the other formats of dynamical systems that we described.

And how about the differential equation-inspired original liquid neural networks? It's clear that the worm is running on a very small amount of watts. Do we, in practice, do you encounter things that are so resource constrained that you have to go to these extreme, extremely specific or extremely biased architectures to actually deploy today?

Yeah, 100%. Think about places where you have to bring in latency to, you know, like computation has to happen like in microseconds. So you cannot really afford to have like larger computational complexity. You want to have the most simple type of system that actually handles that. And, you know, like, there you can have adaptive systems, like, you can have, like, many different formats of systems. And there, this is, like, one place. The other thing is simulation in a whole. As I mentioned, like, in physics, you have, like, physical data that is coming. Like, you want to build a digital twin of a, let's say, you know, like, a chemical reaction that happens at a factory. So in those kind of places, you would go towards extremely biased and even maybe just differential equation-based models like liquid neural networks today. They're getting applied to many, many different applications. I see that because the repository, the original repository from years ago is still open source. People are still building predictive machine learning models on sequential data or physical data or sensor data that it makes sense because the systems are continuous time. And it makes sense to have a continuous time dynamical system to apply to modeling those kind of behavior. And then today with these larger instances like the clouds of the world, you could also direct your agents to actually try out a bunch, like in an auto-research kind of format. You can just direct them to, so, hey, you know what? Go pick the neural networks, the best neural networks that would be the best fit for this type of dataset. And then here is kind of the space of possibilities you want to explore based on the biases that you have. Machine learning people are now mostly orchestrating these automated agent pipelines as we are building. So that's what's happening. In terms of Liqid ourselves, we are trying to contribute to the open source. We are open sourcing some of the instances that are coming out of our, let's say, search spaces, like, let's say, LLM 2. LLM 3 is going to be like, again, you know, the objective is that the next generation should always beat the previous generation on that criteria that we care about, you know, without sacrificing quality. You know, like, that's extremely important for us. And then the range of possible— like, we decide, for example, how large of a neural network we want to make based on these neural networks that we have. And then this becomes kind of the open-source version of our models. Then we work very closely also with silicon companies. We work with AMDs of the world, Qualcomms of the world. And what we do with them, we try to understand the silicon roadmap that they have, the hardware roadmap that they have. And based on that roadmap, we try to kind of inform also what they have to do even in the next generation of their ASICs. Because when we have this understanding of the algorithmic aspects of intelligence, and the variety of things that they could be supporting in order to reduce, dramatically reduce the cost or satisfy the constraints of the use cases that they want to enable. These are the considerations that you have to do. Or even building, like, to the point that we could build also, like, you know, like, specific foundation model graphs, like graphs for them, you know, like, that's— and that's kind of the projects that we do with semiconductor companies. In terms of like on the commercial kind of side of things, we also take models, we apply them in many different— I call the category that Liquid AI is very enabled, we call it device foundation models. You know, anything, all the processors that are outside of data centers, we try to apply our technology to those kind of places. This is outside of data centers. Inside of data centers, we apply to constrained use cases, use cases like low latency, like ultra-low latency applications, AI, extremely long sequences and wanting a very, very small memory footprint of your AI system. You want to have one cost-efficient implementation of these foundation models that you have. We contribute to memory, speed, latency without sacrificing quality and cost. These are the elements that define use cases where Liqid can come in. In terms of access to our technology, new generation of our architecture that comes out of our efforts of, let's say, architecture search or new modalities that we check. For example, with Shopify, we are exploring many different range of applications across recommendations, search, product catalog, understanding multimodal systems very, very well. That's something that we do. Our models are in production, like with Shopify, right now, and they are they're improving the quality of the click-through rates and the criteria, internal criteria that Shopify actually cares about. Our models outside of data centers, they go inside cars. Some of the use cases that we enable is basically in-car intelligence. For example, we power that. Recently, we have a contract we signed. It's actually a historical one with Mercedes-Benz where our models is going to be powering the audio and also the visual elements inside the car. So, whenever you want to talk to your car, basically, the new voice would come out of a Liquid Foundation model, you know? And we control that with a model that is giving you the quality of the best models that you have seen so far, like, let's say, like the audio models that you've seen all around. But at the same time, it is 600 megabytes. It can actually fit inside the smallest processor that is inside that car, you know? So, like, and that, that And that changes the game because you're enabling local AI at scale on places that matter, on cars, on, let's say, on mobile phones, because you can do that. There's a lot of devices. You're talking about billions of devices in the world that mobile devices can be extremely useful. And you've seen that market itself, by the way, just the mobile kind of business is a $500 billion market. This is absolutely insane. And it is as big as kind of the data center kind of market. So, you can imagine there's a parallel here to be made, like the efficient markets and also constrained intelligence market. It's something that Liquid AI is going after. Laptop markets, like, let's say, wearables, robotics, manufacturing, IoT systems, anywhere that we have a processor inside a system, Liquid can bring in intelligence on top of that. So, the aspirational goal that I have is that we really build an intelligence layer On top of the diverse formats of hardware that is out there and processors that are available in the world. You know, that would be something that I would want to have.

I'm glad you said that stat. I had that noted to make a point of at some point. It's worth repeating. The global annual smartphone market is about $500 billion. For all of the hundreds of billions that are going into the data center buildout, and it will get Bigger than the smartphone market, but it's only now getting to the scale and getting bigger than the annual smartphone market. So there's a lot of dark compute out there from an intelligence perspective that is not being anywhere close to maximised. And that does not include the laptop market. So there's $1 trillion, generally speaking, worth of compute going out into the world on an annual basis that are sitting in people's desks and pockets. And there's a massive Massive substrate there to take advantage of.

And we need to do it. You know, like, we need to take advantage of it. I mean, because we don't have enough energy to host it. I feel like we are realizing it right now, like, how difficult it is to really power this format of intelligence that we have at scale. We really have to work smart. We cannot just do the simplest format or maybe more constrained use cases, Let's say if you have a one-shot predictive task or maybe data extraction task and stuff, you don't want to call the fanciest type of intelligence like the Mythos-level model to perform data extraction for you. You can have a variety of different intelligence systems that are doing many, many different things. Then for, let's say, the most sophisticated problems in the world, you can go to the most sophisticated AI systems that are existing in the cloud. So it's inevitable that the world, the processor world outside of data centers should get enabled as we speak right now. And I think there is actually some talks coming out. There's a lag in Silicon Valley, like, in terms of investors that are actually like, just, oh, okay, now that makes sense. There is that lag that is getting answered right now. And I think we are in exciting times also for efficient AI.

In general. Okay. So one challenge obviously with this other trillion dollars worth of compute that's going out into the world is it's super heterogeneous, right? There's many, many different devices, different chips, et cetera, et cetera. So how much of the work that you're doing is about trying to get closer to optimal Use of that compute. Obviously, we talked about scans earlier and the transformers and the data centers. That has been not— I'm sure there's still more optimizations to be made, but a ton has gone into it to make sure that you can get as close to max throughput as possible, right? When you go then do an architecture search and you're targeting some random cell phone or sensor in a factory somewhere, how much do you have to work on like kernels and scans as kind of foundation to even be able to do that search.

Yeah, well, I mean, that's, that's one of the— you see, there's like different layers of abstraction, you know, like, and we're, we're testing out like also like the, the AIs to see like how good they can get like on the kernel design kind of side of things, you know, for the things that they have seen that, you know, in the open source, like the kernels that are available in the open source like this let's say the GPU structures that they have seen out there, but not the NPUs, like the most hidden type of NPUs that their IP is actually not disclosed, those require, let's say, another architecture search because we don't have those knowledge. But I would say kernel design is at a level that we can actually start automating. So you can definitely have loops where you can juice out the best kind of post-hoc optimization. Because some of the things that we were like— whatever we talked about so far has been optimization that we do before design of a foundation model, like all those sort of considerations before we start pre-training a foundation model, right? The approach that Liqwi takes before getting into, let's say, like pre-training a model, we run this massive architecture search before getting that bit of hardware in the loop. But then right after you design the architecture, then there Like all these sort of quantization of everywhere training, changing the bits of the system, like going into the kernel level and try to even juice out even more off of the system. All those post-hoc stuff is also something that is like you can automatically kind of orchestrate it with a kernel engineer that does that and performs those optimizations. And we do that. We do those post-hoc— we do that as a post-hoc optimization step. That being said, we have the capability in-house to define architecture, define operators at the kernel level as well. So for example, you could say matrix multiplication have like on this format of kernels, they are like probably 100 ways that you can actually structure matrix multiplication with caching, with not caching, like, you know, like how distribution between CPU workload and then GPU workload, you know. How do you actually do the entire forward pass as you perform? All of these optimizations could also, in principle, become part of that massive search space as well. These are the questions that some of the silicon partners are asking us. How can we, even before design of foundation model, figure out the computational graph first, but while having the guarantees that when the training is finished, because that's a costly process, multiple millions of dollars has to go into full training of a neural network, even if the neural network is small, you want to have all sorts of inference optimization before the post-hoc quantization-aware kind of optimizations. But yeah, in principle, we can definitely launch something like that as part of our search.

So what do you think hardware makers should be doing differently? What are you telling them to prioritize? to help us realize this more efficient distributed AI future?

I feel like with every iteration of a technology, software technology, like the level of abstraction that the hardware providers should be building for it is coming up. For example, we've been talking about CUDA mode for a long time now. There is no CUDA mode anymore. Why? Because if you look at the AMDs, there are agents at the kernel level, at that level, now they can automate that stack. The stack that foundation model companies should— sorry, hardware companies should start working on is the intelligence layer, like what NVIDIA is doing with their Neutron project. If you see how this project is actually quite successful, and probably a couple of billion dollars actually went into design of these Neutron projects. And if you really look at what does it do for NVIDIA, it's just like basically building an intelligence layer on top of NVIDIA's compute so that you have like an easiest barrier of entry as an enterprise to buy NVIDIA solutions, you know. So that's the sales of the solution is always like around those things. So from the— from it, I'm just talking about the business of like enterprise business of these hardware providers, you know. If they want to sell more hardware, you know, and they, they're gonna up their stack from that kernel-level optimizations. Those are kind of post-hoc things, and they get into the intelligence layer. Now, does that mean like they have to become a foundation model company? To some extent, yes. And they have to be able to train that intelligence layer themselves. And NVIDIA is like a success example there, how, how this thing is actually like, you know, paying dividends for them, you know, like building the Nimotron project. If you look at the other foundation model companies, like the other hardware companies, they have not done this yet. They have started strictly optimizing their models for the open-source architectures that are out there. That's also extremely valuable. You have to do that. That's a given. But again, if you want to create a differentiated value and you want to be successful, I would say you've got to be able to bring the stack to the intelligence layer. And that intelligence layer should naturally fit on top of your hardware. If your hardware has limitations, let's say in terms of to your competition. Software can always help to give you that juice. Like, for example, if you're talking about maximum token speed on an AI PC between an Intel computer, between, let's say, an AMD computer, and let's say a Qualcomm computer, they might have like their own trade-offs. And the, the company that is going to have an edge is a company that actually owns that efficient intelligence layer on top of their hardware so that they can say At the end of the day, you're going to run tokens on top of this system. You know, it doesn't matter what's the bandwidth of my, you know, CPU, if it's like a little bit lower than you or higher than you, or the amount of memory that I'm actually having in there is lower or higher. With the intelligence layer and software optimization kind of techniques, I can actually get myself to the place where I'm actually the winner here. And for example, all of my laptops or all of my PC solutions that I'm actually building as a hardware provider They are naturally coming with an intelligence layer on top of that. Nvidia is entering this game like also heavily, you know, as a competitor in the CPU space now, you know, as you're seeing like devices like Google actually entered with Android, um, with the Android ecosystem that they have, like they, they're replacing Chromebooks with Androids, right? Like with Android like laptops, you know, like they are— this is kind of what they're, what are, uh, what are putting out there. I think it's called Aluminium OS or something like that. I don't actually know what exactly it is. And then Meta is going to come with a sort of devices themselves as well. So all the processor builders, I have a feeling they need to get a little bit closer to that intelligence layer, you know, and try to leverage software and efficiency kind of considerations into account when they are going for their, for their planning, you know, like planning ahead, you know, where things are going and why that intelligence layer is important. Because right now agentic AI is something that everybody else wants. All that applications that you want to do on silicon, you want to do that on a base of intelligence that is already provided to you. You know, you want to build harnesses, you want to build like all sort of— the harnesses are basically applications that you can build for solving a certain problem in, let's say, in a traditional sense of an application, right? And, and that, that market can get enabled on top of chips if you have a very nice stack that enables intelligence on top of what you have. The other thing that I would say is that silicon is always— if you're going outside of data center, I mean, there is a massive diversity of silicon, but then the resources are constrained no matter what. You don't have that much power. So you got to be very careful what sort of applications are the most important uses of your hardware medium and platform. And then Those use cases, if you already have an intelligence layer, you could identify when is the time to sunset basically a set of, let's say, devices. For example, I strongly believe— I think glasses could be like a format of computation for hardware later on, like devices. I think glasses would be like very, very interesting kind of medium. And I think somebody's going to get it right eventually. And it might actually be a replacement for, let's say, for laptops.

Right.

And then, but is it? We need to really figure out what is it. But if you really want to define a market and define where the field is going from a hardware perspective, I would say you got to up your software game by a lot and really get hands-on and investing a little bit more energy and capital into the R&D side of things on the foundation monosight. Because that's the new base of software, inevitably.

Does that imply a future where we have just a lot of vertical integration and a lot of coupling? The model will come with the hardware that I buy, and it may not be so swappable in the future because the model is heavily optimized for the hardware and vice versa, such that these things are not so modular in the future as they are today?

Or why do you want to change the model? It brings Yeah. You're given basically a default. Now, you want to switch this thing, by all means, but why do you want to switch it? If this model, the intelligence layer that is in there, it's not fixed, it's an adaptable system, it is a, let's say, self-improving system, like with a call to your datasets and stuff, you can actually have a platform that does, let's say, full fine-tuning of that system. It is enabled. there is not just one model that you can load into your system, and there's not just one cloud model or one on-device model that you're going to use. You've got to be able to orchestrate between, let's say, many different instantiation of this model to be able to build application. We're talking about a model class that would actually sit on top of, let's say, hardware on a laptop, for example. Yeah. I would say you need to have that, but if the default is just giving you that efficiency and that it's ready to go, that's the choice. That's something that I think NVIDIA is trying to propose in enterprises when they go and sell the NVIDIA Neem projects. When they go out there and sell these things, it makes a lot of sense because you already have a project, you already have a multimodal model that is loaded on top of your, let's say, the PC that you bought. Why should I switch? Because they already have done all optimizations for me. You know, and it is running like extremely fast. Why do I need to change it? And the model itself is tunable. I can use their Megatron kind of framework to actually tune the model. Now, if you don't want to do that and you want to just choose like another model to host it, as I said, this has to be given. You know, you have— your hardware should be already— you have to be optimizing for the entire open source ecosystem and, you know, like models that are available. But at the same time, it would give you an advantage, like, you know, like To yourself and to your customers.

Okay.

Maybe in the last few minutes, how about a practical application of this? You have a blog post on local co-work, no cloud, no waiting, tool calling agents on consumer hardware with LFM2-24B A2B. So that's a, I think people know that's a mixture of experts, 2 billion active out of 24 billion total parameters. Let's say I want to make that a part of my life. I am interested in how you would coach me on setting this up. Today I have, for reference— and I, by the way, try to make the— sometimes I try to make the transcript of the podcast something that I can feed to my agent. So you can think of this as partly coaching me, partly coaching my agent. So I've got this deep context database that is the last 5 years of my digital output. This podcast will be recorded, transcribed, obviously recorded, transcribed. It'll go into this database. So everything you said, everything I said will be in there and searchable. And it's got all my email and Slack messages and everything. Okay, cool. So now I've got Claude on my desktop that can call tools locally to get data back, but then it sends all the results to the cloud to decide which of those results are actually the right ones to be looking at. And so far I've been okay Okay with that. The benefits are certainly worth whatever risk I'm taking, I feel. But I would maybe love to run that data through a local model first so that I don't have to send all my data to the cloud every time. Now, notably, what it gets filtered through, I'm probably still going to end up running through a foundation model, so it's not going to entirely skip the cloud. But also, I'd like to save some tokens too, because I'm going to be wanting to use Fable whenever I get it back for whatever its most appropriate So how do I get really good performance on my local computer with this model? Do I need to be doing fine-tuning? Do I need a distillation strategy? How do I actually take the base model that you have and recover as much of, let's say, Opus or even Fable performance as possible? in terms of searching through, understanding my data as I possibly can? And how much is that? What should I expect in terms of how far I'll be able to push that process? How close to parity with frontier models can I get? So tell me everything I need to know and I'll go have my agent do it.

Yeah, well, I mean, that's a great question. Obviously, the local co-work is, as it stands today, obviously just to open minds. hey, you know what, this type of applications you can enable. The class of what this format of local agents are going to enable is basically a local computer. You have an orchestrator. If something is so complicated, it should be able to send it to the cloud, fetch the answers for you if it's not sensitive. If you have, let's say, smaller models that are PII models, it should be able to use the PII models to really filter out all the personally identifiable information information, send it to the cloud for you. You don't need to even see these models. These models should be on the background. The model that you have to tune is that orchestrator that'd be able to route between many different services or even smaller specialized models that are doing stuff, and also some of the cloud models that are out there. That router is the computer. That's the local computer. That's the definition of one. When you open your laptop, it should be just that. You literally have just that. and then you start like working with all the services you want to have. It's just the same way that you communicate with your, let's say, assistant, you know, like it should be like in the format of an assistant that does all sort of those jobs, you know, like especially it takes, it takes a while for people to get, uh, get, uh, not weirded out by the user interface, like just being that and not seeing like all the like file formats and you know what, what, what you have to do, but you're gonna get used to it, you know, like now Thoughtcode is like even as for the ID. Now, is the 24B off the shelf going to be there on the quality that does all of those stuff for you? No, it's not. None of the local models today are there. On the local models today, you got to fine-tune them. You got to get them to be specialized for the stuff that you want to do with the proper explanations for your Claude agent. You basically give it these things. But is Claude being able to go out there and actually build this model for you? Let's say do the fine-tuning and getting it to that place? No. Because today, even FAIR level models, they would not be able— first of all, you wouldn't have access to that because Anthropic is actually having access to the auto-tune and stuff like automated models generating models platform. There are companies and also ourselves, we are building platforms that enables you to do fine-tuning of this whole thing that would cost you between tens of dollars to, let's say, low thousands of dollars to actually get you to the cloud quality for your— and with all the checks, production quality model. It's not going to cost you tens of thousands of dollars. It's going to be between tens of dollars to actually those thousands of dollars. That's the scale that we are thinking about because, again, efficiency here matters a lot because we don't want— We want this fine-tuning to happen on the compute that you already have, or if you don't have it, you would actually put it in a secure and proper data center that you would actually host it from the providers basically, and you would be able to get to that quality of the models. There are platforms that hopefully as we go forward in the next few months, we're going to announce some platforms that you can hook it just directly in your terminal. You don't have to do anything. You just say, hey, you know what? Go call this platform. Now for fine-tuning this thing, go call this platform. This would give you a production-grade foundation model that you can deploy it now for yourself. Either fine-tune on that data or depending on the use case, depending on what you want to do, it doesn't even have to see that data. It can even synthetically generate data on its own and actually train the model to be just the perfect reliable tool caller and understands when is the shortcomings of it and it can actually go away. And deliver it to some other places. Yeah, so I would say we're extremely close to get to that point.

I'll wait for your platform to be ready. I don't have to DIY it, is the message.

Yes.

This has been brilliant. Maybe one last question, and then I'll just kind of give you the floor to close however you'd like. How far do you think this goes in terms of kind of miniaturization of intelligence, if you will. One intuition would be like the biological world that we see is maybe on some sort of Pareto frontier already, and so the watts that go into our brain is maybe like getting us close to the maximum that we could get for that amount of power. But maybe you have a different intuition. What would you expect in terms of upper limits of intelligence that I could have on my phone or on on my laptop as we really get to the kind of physical limits of the technology?

You know, when you think about intelligence, like the way I look at it right now is that like transformer-based networks and also like our type of, let's say that the current architectures that are landscape of architectures that are available, they gave us with scale, they gave us in-context learning capability. The thing that actually emerged from next token You see, the word is important— emerged from next token prediction. Intelligence for me is an emergent property. If you want to get into— like, if you want to miniaturize kind of intelligence, bring it like to the physical world, I don't believe with the current set of algorithms, like, we would be able to get close to the, let's say, intelligence per watt that human brain is actually like providing, right? You're not going to get there.

Why?

Because I believe Human brain over the— we also have to consider the amount of energy that went into design of humans as a whole. Biological evolution is actually a very, very long process. A lot of the pre-training process that people say, oh, a lot of— you're going to see the entire world. Humans do not need to see the entire internet to be able to reason about something. No, but humans have gone through years of evolution. I would actually attribute it a lot to the evolutionary aspect of But again, what I would say, human intelligence came with multiple mechanisms of in-context learning. You just don't have— when our current AI systems do in-context learning, they learned a vague representation of one algorithm, which is least squares. It's basically least squares. So, what they figured out is basically gradient descent in a mushy way, like for some use cases, With some examples, you would be able to make the system understand and give you the next example. These are beautiful properties. This is what I would call emergent property of these systems. You set the algorithm to be next token prediction, you got a vague version of a gradient descent in context learning capabilities. For humans, you don't have only emergent gradients, you can learn by examples, but you can also do reinforcement learning. You can run simulations in your head. You can do all sorts of algorithms. You can do Bayesian statistics in your brain. You know, you can— you, you see what I'm saying? So you have like a diverse set of algorithms emerged from the way that humans actually got designed and got intelligence. For me, from— I, I believe if you start optimizing for those, if you try to force your way into the system to become a reinforcement learning learner, but learning from trajectories, that's not the right way to actually get to emergent property of intelligence. I believe the format of intelligence that is going to be miniaturized, you know, like, like being like in a smallest amount of time, maximum amount of intelligence, like the human brain, we need to come up with these identifiers, like these confounding kind of algorithms, which is next token prediction is one of those. What else do we have to do at the beginning of the design of these systems so that reinforcement learning, curiosity-driven intelligence, drives with limited amount of energy from that final system that actually comes out of it. So I would say that emergent property is something that we need. There's a lot of research that has to go in that direction, I would say, and multiple variations of ways of learning that would enable the next generation of artificial intelligence, I think.

That could be a great note to end on, but let me just Give you one more opportunity. Anything else that I didn't touch on or anything else you just would want to leave people with before we break?

No, I think we covered a good lot of information. I think we went into some things that I don't usually talk about. Now, in my current role as the CEO of the company, I get to talk a lot more about the business opportunities. Opportunities in the market aspects of the intelligence and stuff, but it's still like you know at the core like we are scientists and we are really like pushing the boundaries. Like we are actively thinking about it day to day. You cannot imagine like even right now if you're talking like I'm already in in the backend like I'm actually like training some stuff and and and I really don't want to get away from from that kind of mentality because I think everybody can actually be able to now build you know like and I think that. The times couldn't be better. The agents are pretty amazing at giving you this opportunity of really doing frontier-form research. If you're curious about something, it's just that we need to find people that are a lot more curious and improve the curiosity in people as well to really get instead of just fear of everything. I'm a techno-optimist. Probably you've observed that. I usually talk about technology. really in a very, very positive light. And I want people— like, one thing that I always say is that we need to really get to the level where people are as curious as the scientists were, like, from day one. Like, what gets us, like, into science and the purpose of science has always been satisfying our own curiosity, you know, and also understanding the world around us, right? So that's kind of the purpose of science. And I feel like that, with the current superpowers that are given to us, like, with these AI systems, I think everybody could would actually start really contributing to understanding the world around us and also generating more value and also just satisfying our own curiosity. But also, it requires a certain degree of rebasing our biases towards, oh, what is work, what is not, and what is automation? What is an army of agents working for me? What's that definition even look like? And then changing culture and changing way of thinking about this thing, not just for individuals but also for enterprises. we are dealing with as well as Liquid AI. And yeah, so I think that change is something that I'm looking forward to. And I think this is something that's going to happen. The pace of it is extremely fast, but I think we got to do it one way or another.

I love that. I consider myself extremely fortunate to live, to a really remarkable degree, a curiosity-driven life these days. And that's one candidate for a positive vision for the future that I think I think should inspire a lot of people, especially because, as you note, there's an opportunity to move into that phase already with the AI systems that we have. This has been excellent. I really enjoyed it. Thank you so much for the time. Ramin Hassani, co-founder and CEO of Liquid AI. Thank you for being part of the Cognitive Revolution.

Thank you.

Cut it down, cut it down till it's clean and bright. Big ideas, small machine, just a new design. Feel it, move, feel it, but never frozen in time. Flowing to your hands, falling on the land, up the cloud and to the For every little thing, just enough to make the whole horizon sing. Cut it down, cut it down till it's lean and bright. Big ideas, small machine, just a new design. Feel it move, feel it bend, never frozen in time. Liquid flowing through your head. Where's it going? Everywhere we'll be in the car and upon in the open air.

Cock cock cock cock.

Meet the revolution, intelligence everywhere. Feel it coming down, coming down slow. Up the cloud and to the ground.

If you're finding value in the show, we'd appreciate it if you'd take a moment to share with friends, post online, write a review on Apple Podcasts or Spotify, or just leave us a comment on YouTube. Of course, we always welcome your feedback, guest and topic suggestions, and sponsorship inquiries, either via our website, cognitiverevolution.ai, or by DMing me on your favorite social network. The Cognitive Revolution is part of the Turpentine Network, a network of podcasts, which is now part of a16z, where experts talk technology, business, economics, geopolitics, culture, and more. We're produced by AI Podcasting. If you're looking for podcast production help for everything from the moment you stop recording to the moment your audience starts listening, Check them out and see my endorsement at aipodcast.ing. And thank you to everyone who listens for being part of the cognitive revolution.