Welcome to the Practical AI Podcast, where we break down the real-world applications of artificial intelligence and how it's shaping the way we live, work, and create. Our goal is to help make AI technology practical, productive, and accessible to everyone. Whether you're a developer, business leader, or just curious about the tech behind the buzz, you're in the right place. Be sure to connect with us on LinkedIn, X or Blue Sky to stay up to date with episode drops, behind-the-scenes content, and AI insights. You can learn more at practicalai.fm.
Now, onto the show. Welcome to another episode of the Practical AI Podcast. This is Daniel Whiteneck. I'm CEO at Prediction Guard, and I'm joined as always by my co-host Chris Benson, who is a principal AI and autonomy research engineer. How are you doing, Chris?
Hey, I'm doing great. Can't wait to get into today's conversation. It's going to be fun. Yes, yes. For an audio podcast, we're going to talk about a lot of interesting visual things.
Maybe before we get started, just a little teaser, Practical AI is posting some videos on YouTube now. So, if you do consume podcasts that way, you might go check us out on our YouTube page. But speaking of images, videos, and more specifically, image generation, really excited to have with us today, Dustin Podell, who is co-founder and researcher at Blackforce Labs. Welcome, Dustin. Yeah, thanks for having me, guys.
It was really great to be here. Yeah, and I know that Blackforce Labs does more than just raw image generation. There's a lot of workflow-related things, hardware optimization, all sorts of cool stuff you're involved with. But as we get into some of that, I'm wondering if you can just help our audience with a bit of a state of image generation methods and workflows for the industry. We've talked on the show before about diffusion models, and we'll link some of those episodes in the show notes maybe.
But a lot has happened. There's a lot of people working on a lot of interesting things. And I'd love to kind of understand over the last year, what are some of those main points that might be good for people to orient themselves to where things are at now? Yeah, no, it's a good question. If I'm allowed to take it even a little bit further back, I mean, the state of...
Absolutely. Yeah, yeah, the state of the state of the image gen, video gen, generative models as a whole has kind of gone crazy, so to speak in the last three or four years. So where were we now? Where we came from about four years ago, where I first entered into the scene, so to speak, is we were at models that were essentially just doing little blobs of color that were kind of related a little bit to where you were with the prompt, you know, okay, oh, a lighthouse on the beach, and that. And you would get something that, okay, that vaguely looks like it, and you would show someone in the...
Yeah, I can see it, I guess, and maybe it interests a few nerdy people. And then now today, we're at the point where, you know, if you've probably been seeing all the plenty of the stuff online where we're seeing whole short films made entirely with AI generation, where certain scenes are almost entirely indistinguishable from reality, so to speak. So I would say we've come quite far, but I will also say the core of the technology hasn't actually changed that much. It's been a pretty nice, like, forward progress. I don't want to, like, dive immediately in anything technical here, but it's certainly, I mean, for anyone who's been paying attention, I'm sure, or anybody who really hasn't been paying attention, this probably came a bit out of nowhere.
So, you know. Yeah, I was going to say, I guess, you know, with the broadly attention by the general public is so much on kind of more of the LLM generative world in terms of, you know, that stuff, and everyone's finally on apps, regardless of whether they're technical or not. And so I think a lot of people kind of miss the tremendous advancements you guys are making on that side. So, like, you know, could you take a quick moment and now that you've kind of done the highest level, maybe step through some of the things, and people may remember, like, we talked about stable diffusion and things, a couple of those things that have kind of led up to what we're going to dive into today with some more specifics. And as Daniel mentioned, we can offer some links for past conversations if people want to dive into those specifically.
But that would really be interesting to kind of hear distinct points on that timeline as we hit to this point. Sure, sure. Sure. First before I do that, I guess I'll ask Chris, how technical am I want to get here? You're a lab.
So we'll, like, if you dive into something, we may stop you and say, that is what is that acronym? But other than that, you can go with technical as you want, and we'll make sure that everybody can stick with us. Okay. Yeah, I guess if I'm allowed, I'll do kind of at least the best of my knowledge overview of the last bit of time. So obviously, the field of AI on ML has been around for a while trying to kind of structure and learn distributions of different data, mostly with use for tabular prediction and all sorts of other things for a while.
And then in 2017, some researchers at Google figured out this very cool thing called the transformer, which I'm sure people have heard many, many times about right now, which I guess the best way to go down this path is to say, like, we found a way to train much more general models, like learn from much more general data. I'm trying to think how to essentially take this story. Okay. It's a dramatic pause. It works.
Yeah. That's good. Yeah. Essentially, the story I want to tell is the story between the kind of, I don't want to call it the battle, but that's been going on last year's, but essentially the story between diffusion and autoregression that's been going on the last year's. So let me start by trying to define a little bit with this huge autoregression is autoregression is anyone who's been playing with language models?
Oh, you know, all these, you know, GPT, Claude, all the new stuff, everything that's been going on last year's, these are what you would call an autoregressive language model. And so what this is, is it's predicting data essentially one piece of the time, so it predicts one piece of data, and then it looks, okay, what did it make? Then it predicts the next piece of data, it looks at what it makes, then it predicts the next piece, and it slowly, slowly, slowly, slowly, slowly, it builds out, you know, at first, it was just very distancing language, and then it became, you know, now I mean, we're having agents and a lot of very, very fancy things with them. On the other side of the coin has been something called diffusion modeling. And I mean, I don't want to get this to you because now we're doing more like flow matching, but let's go with the fusion as the, as the base term here.
And with diffusion, what you're doing is you're not trying to model one, say like in language model, it's, you know, you call it a token, but you just do like a one word or the time. With the fusion, what you're trying to do is essentially, trying to think how to tell how to tell how to tell a story. It's all good. Yeah, yeah. It's all good.
I feel like I'm watching like a thriller where you're about to like reveal the next thing. It's good. Yeah. So what we do with diffusion is I'm just going to tell practically what we do, and then we can kind of break down, you can ask me a question. That's fine.
So what we do with diffusion is we take a whole continuous medium, and what I mean by continuous medium is the most famous one is images. The world is continuous. It's not, you know, it's not that there's, you know, one, like a letter is disparate. There's a T, there's A of E, C, A D, and E and F. A color shape the structure of the world, the color, you know, the light, all of this, this is this, these are continuous natural mediums.
And so what we do is we take one of these continuous natural mediums that we've now encoded in a digital space. So we use RGB. So, you know, okay, 256 colors we typically use for red, green, and blue. And then we encode this into an image. And then what we do is instead of trying to predict, so an autoregressive you might try to predict like it's like pixel instead, what we do is we essentially try to remove information and the way we remove that information is by adding noise.
So you imagine you like add a little bit of grain on top of an image. You can still kind of see the image. Maybe it's a little grainy. Maybe it's a little blurry. Oh, you can't quite see it.
And you got a little bit more. Maybe you see some of the shapes. Maybe, okay, there's like, is that a dog? Yeah, you kind of see it. The color starts to fade.
And then you keep adding and eventually you just get, okay, this is just noise. I don't see anything in this image at all. And with the diffusion model, what we're trying to do is we're essentially adding a bit of noise. And we're saying now try to predict what the clean image is essentially for this. So, you know, okay, so we had a little bit of noise, predicted clean.
It just has to remove a little bit of noise. You've got a lot of noise. And now what it has to do is there's so little information that has to actually like come up with essentially like how to infill this properly. And then what you do is then in what we call inference, which is when you actually run the model, you ask for an image, what you do is then you start a fully noisy image and you say, give me a dog, you know, wearing a top hat on the beach. And then essentially what it does is it tries to remove a little bit of that noise to get closer to this idea of a dog in a top hat at a beach.
And you might get some structure. It tries to figure out its very coarse. It's very just trying to figure out the shape. And then it has a little bit of shape to latch onto. And then it does the next step where it removes a little bit more of that.
Oh, and it starts to come into view. It starts to get a little bit more in focus. And you do this do do do do do do do do. And what you see is essentially you completely remove the noise of this image. Oh my goodness.
Now you have this completely generated image of a dog in a beach with a top hat. For a long time, this was a, I don't want to say not good. I mean, all the models have gotten a lot better. But when it first started out, these look like blobs of color. It was like these little like you could barely see it.
But fundamentally what I just described is still the process that's been happening in the last four years from these old image generator models or anyone that maybe tried like Dolly Minium in the old state of the fusion models that we work on. You know, I don't know. There are many, many of the different models are out there now to these modern video models that are producing whole 10 second sequences. Fundamentally, it's doing the exact same process of removing this information slowly with this noise and then just slowly going the other direction, essentially creating info to info this. If this makes sense, I'll pause and I'll pause and I'll just kind of like.
So let me tell you, because as I'm listening, very, very good explanation, probably the best one that I've heard. So you are right on target. If you just carry on what you're doing. So I'm enjoying this. And you mentioned that when this started, right, you sort of ended with these blobs, et cetera.
I think there's a lot of people that have seen very impressive things like the, like you were saying like whole parts of maybe movies or commercials or something being generated in this way. What is the current, I guess, state of the art in terms of whether it be your models or others. And I know there's more to talk through as you get in, like there's more than just raw image generation. There's how this fits into workflows and doing certain tasks. But what is kind of the state of the art in terms of what we're able to generate?
Where are the bounds currently and in terms of quality and that sort of thing? So I'll tell you kind of like in the last year, I'll split it up into, let's call it three different categories, but they kind of blend a little bit, which is these continuous mediums I talked about. So I gave the example of a new speed with image. Video is obviously an extension of that adding the time dimension. But then there's also audio as well, which now a lot of these video models generate audio.
But then there's also things like I don't know maybe for the like, sooner where these music models, some of these are also using this exact same technique of removing information with noise, bringing it back in. It's just a different medium that they're essentially applying to. I'm happy to talk about our models all day. I'm talking about our models. But I also want to be fair to just anyone listening and honest to kind of the state of the world on if you want to go out and try some nice things out there as well.
I will say that before I dive a little bit into what I think is maybe the best one to recommend, I will make one statement of that like, best quote unquote here is a little bit hard to define sometimes. And this is something we're always wrestling with is what makes a best model for someone. Because obviously you, when someone uses these models, they're typically prompting or putting in now you can reference images or videos or other sorts of things and kind of treat it a little bit like a creative companion. And then what you expect out of this can be very different for different types of people. And then I would say that if you want to see kind of the quote unquote state of the art right now, for what I would say is like text a video at the moment, it would be C-dance.
So C-dance has done some extremely impressive work with the C-dance two model over the last year. So they're doing, I think it's now 4K generations up to 15 seconds, which is very nice quality. It's focused very much on like cinematics. I will throw a shout out to anyone that, you know, if anyone from the sort team are listening to this, I think sort of was quite a nice model. It didn't rank so high on some of these like leaderboards that we, in the AI community, there's a bunch of, I'm sure you guys talk about this one, a leaderboard rankings where models place.
I think the LLM side of things is much better covered where there's very clear metrics of how well does it program, how well does it do math. On the more creative side of things with video models, image models, audio models, we typically end up falling down to this single preference type benchmark, which is just like a general preference. Everyone in the world goes and can can vote on, you know, on these leaderboards. It's a couple of different nice companies and sites that have to do this. But I feel like this kind of removes some of the specificity that the granularity, that's the word I was looking for.
Listen, I've been to an incredible amount of AI events, many of which are good, but many of which are not practical. You know I love practicality, we're on practical AI, and some are just hype focus, some are sales focused. That's why I'm always eager to share about an event that I truly think is practical and useful for people. That's what I discovered last year at the Midwest AI Summit. And they're going to have another Midwest AI Summit October 15th in Indianapolis this year, 2026.
One of the reasons why I love this event was there was an actual AI engineering lounge where you could sit down and talk through your use cases with actual experienced AI engineers and practitioners to really brainstorm and come back from the event with actual solutions and practicality rather than just a bunch of content and slides. But there were also amazing keynote speakers, speakers from even that had been on the podcast before like Regeve Shaw was at last year's event. I would really recommend that you go to Midwest AI Summit.com and for our listeners, you can actually get 20% off with the code practicalAI20. So go to Midwest AI Summit.com, don't miss your chance to attend this event and get 20% off with the code practicalAI20.
I was going to ask you, are the models that you're describing, are they kind of the traditional diffusion models or are you mentioned in passing a few minutes ago about flow matching? And so I was just in my head, I'm trying to kind of categorize how do the ones that you're talking about right now, how do they fit in and how, what is it to go back and pull that up as you were talking about diffusion and you made the reference to flow matching, what is that transition and how do those models that you're talking about now fit into that? I'll first keep it very simple and say all of these models are doing the same process of what I described earlier of adding noise and training the model to then remove this noise. The difference kind of, this is where I guess something more technical in how we actually approach this.
So we used to do more of this process called diffusion, which I don't know, I don't know if I want to get into how deep and technical this is, but essentially what we've done is we've cleaned this process up to what we call flow matching. And flow matching is essentially this very simple process of still doing the same thing, still training the model to remove this noise. But fundamentally what it's learning under the surface is anyone out there who has ever seen like a velocity map or like a flow map, what this essentially looks like is if you can envision like a, I'm just wondering if the whiteboard behind it. For the audience folks only. It's about to turn to the whiteboard.
Yeah, it was great. Dustin, if you're on video, you saw it, Dustin was literally about to turn back to the whiteboard behind it, which I love. If I was in the room, I'd be like, go man, take me there. Unfortunately, I'll give it to the audience. We'll be able to see it.
So you're going to have to describe it. Yeah, yeah, no worries. Let's keep it purely in the auto space. So I'll draw a picture with my words as best as I can. I spent enough time prompting.
Anyway, so I can do this. All good. But yeah, essentially the way I like to think about it is, if you imagine like a, a, a, a, a landscape, like, I don't know, you can imagine your town or the region you're in and imagine you're looking at it from like the sky and you see like your house, you know, maybe right in the center of this map. And what you might want to do, now, now imagine, okay, there's wind going all over the area. And the thing that you want to do is you want to be able to throw up like a paper airplane from anywhere in this, in this landscape or city and you want the wind to carry it so it lands on your house.
And functionally, what we're trying to train here is essentially that in a much, much grander hyperdimensional space where instead of it being your house and instead of being a wind in a paper airplane, what it is, is your house in the scenario is what we call the manifold of real images. It's essentially the place in, I'm trying not to use to me, stop me if I used to me where it's here. I'm going to say I'll ask you, but you're fine. Keep going. We're fine.
We'll just get explained it as we go. Yeah. It's the place in latent space that would be where the actual real images, images are. So, um, another way I'll, I'll try to describe in a simple terms. Okay, with your town here, you're in 2D space here where you are above an x, y and the paper airplane has to fly in this x, y coordinates.
So you land and then you have to x, y in this space, instead of it just being too coordinates, it's enough to have all of the colors of the entire image. So that's why I say a type of dimensional. It's, it's the center is where literally real images are. And so what we're doing is instead of it being wind in the rest of your town, imagine the rest of your town here is every image that isn't real. And what that means is noise.
If you think about noise as a real image, you can go generate a bunch of noise and say that's your computer, but that noise would exist somewhere in the space of all possible images. They're not real images. They're, they're noise, but they, you can save them, you know, they're, they're, they're RGB values. You have them. And so what we're doing with this, essentially this flow matching is we're training the model, like when we say we're, we're training it to remove noise, what's really happening under the surface is we're training this flow map so that we can land anywhere in this field of noise.
And then these flows, these winds in our scenario, when you take a step, we'll take you closer to your house or to the manifold of real images. And that's what like, as you get closer, it starts to look more and more like a real image. It's literally has this. And then we actually finally land on this, this manifold, boom, you have, hopefully a real image, but you'll have some of it, you know, and then the better the model is trained, the better you have mapped this flow to actually take you to where you want to end up. And I'm assuming, yeah, no, and I'm assuming because you are kind of mapping this from, from where you start to where you end up with this real image, maybe in a way that's just to be crude, I guess, less, less random.
I mean, at the end of the day, all AI and machine learning, like it's sort of that training process is very much trial and error, but we have a lot of optimizations around it, right? So I'm, I'm assuming that this then allows you maybe to, I guess, advantage wise, shorten the training or make that more efficient, or am I, am I misconstruing that in some way? Yeah. Definitely like, one of the things that's taken us a lot further over the last three to four years is just figuring out a lot of optimizations, both in the actual training itself. And then also in just, you know, architectures of the actual models have improved.
But I will say the like fundamental underlying process of just removing this information and finding a path back to it, it's still the same process. It's just like the car, you know, you've been having cars go down the road since the 1950s and we've made better engines and better safety and all this, but you're still driving down the road. Like that, that's, yeah, that makes sense. And now definitely, I think that's a good setting in the sense that we've got to a point where, you know, the models coming out of Black Forest Labs, the models coming out from other places, the models that are even like now in my text messaging, right? I get an image, I can immediately remix it with an image model, right?
And so I, so these things are becoming more embedded in our lives, but I don't know if everyone in the audience, some might have been along for this ride where I was like, oh cool, I can generate an image of an astronaut riding a horse on, you know, you know, wherever. And that doesn't seem that practical to folks. So I'm wondering if you could now kind of, given that foundation that we have, and we know sort of where we're oriented in the state of, I love how you put it on your website, Visual Intelligence, which I think is, I love that statement because it gets to more like language, although it's being used a lot in terms of agents and intelligence, like you mentioned, it is, it is very much a subset of the information that we process as humans, right?
There's this visual element, there's audio, et cetera. So now that we have that foundation, could you help the audience understand some of, I guess, the practicalities and the outworkings of the like, okay, we can do this now. So what? So how does that, how does that help people in the real world in ways other than maybe just pure creativity? They're certainly like the cinema, like you were saying, that side of things, not everyone's going to be generating movies, maybe, or maybe more people will, I guess.
But yeah, I think you're understanding what I'm saying. Like, where is this going to impact, where is it impacting me now? Where is it practically going in terms of the application? Yeah, yeah. I'm very happy to talk about this. This is a, I think we're going through like a very
nice transition period right now where we finally get to leverage these for some very useful things. And if I'm allowed to take a little bit of a tangent and kind of, yeah, again, wherever you want to go, it's good. All good. Yeah. Yeah. So I mean, I guess I'll take it back to like a little bit of timeline
of things. So okay, so early on, we had these nice models where most people recognize them for you put in a prompt, you get an image out, you put in a prompt, you get a video, maybe you get a song, it's just this one way, okay, you make something, this appeals to maybe creators or add agencies or movie, you know, now with cinematographers and adding, we love this stuff. We love the creative side of this and what it allows because it fundamentally to me, this is like a nice potential communication tool. But then I feel like something changed. I don't want to say change, but there was definitely a timeline moment when we started to move into editing. So was it two years ago, a year ago,
an already year give or take, we released our first in context editing model called Flux Context. And on the surface, you look at this and go, okay, well, this is this is an image editing model. Something you can take a photo, you can clean it up, you can add a hat, you can do silly things, you know, you can, whatever you want to do, it's a general editing model that's supposed to do all these in our relational things. And on the surface, that's very cool. It seems like a creative thing.
But if you think about what's actually going on under the surface for the model to be able to do this, it has to understand a significant amount of relationships in the world and what it means for these like, like how these relationships actually interact together. So if I say take a picture of, like a, I don't know, we look at water glass here on the table. And I take a picture of water glass, and I say to the model, not the water glass over and shows, you know, show me what happens. The model has to understand some part of the actual world. Like, it has to actually model the world in some way to know, okay, it spills over, maybe something gets wet, maybe x, y, z happens. And
essentially, what we're trying to do is train a model that could do all of these types of relationships. So the model is learning not just like this one thing, but just how the world works fundamentally. So you can do this editing. Now we can take this a further step further and we can look at all the video models that are that are coming out right now. And you can do a very similar thing. You can take,
you know, in an image and say, okay, this person now grabs a fire extinguisher and puts out a fire and it has to actually understand these relationships to do that. So well, maybe this wasn't our original goal way back in the day. You know, we were trying to make very cool models and to go ahead. I mean, I think it's sometimes we want to model the world that maybe weren't thinking this far ahead. Inherently through this whole process, we've learned to build these models that are developing this and I'm trying really trying to avoid the use of the term world model here.
Because you've got to go there if you don't. I'm just telling you. Yeah, you might notice me trying to like skirt this because I think it's a little overused these days. But we've been talking about it. But this is many people, which is where I was going.
Yeah, yeah, fundamentally this is what these models are doing when you train them at scale and you really train them to be general and robust is they need to be able to simulate parts of the world to get an output. And on the one end, you can take that and go, okay, well, this is nice for creativity because you can make a film scene and it looks nice. But on the other side, this is why we're starting to move in this area, you know, this area we're holding digital intelligence and now we're starting to see how we can actually leverage what we're calling, not just us calling it, this is a general field term, but the representation inside of this model of the world to actually go do practical things. Now, this is to say, we're going to talk to the creativity side of things. We're still definitely pushing on the side. This is something that we're also very fond
of. But looking forward, like if our models are understanding these relationships in the physics of the world like this, well, this is a great place to put it in, say, robotics and actually, okay, now a robot has this understanding, this model of the world inside of it, to go act and take actions with confidence in the world. So, Alita, there is kind of a kind of trying to build up to it. Let me ask you're in the area that I spend all my time, which is embodied intelligence, robotics, UXVs and stuff like that. That's my world. And so, even within our space, the notion of world models,
there's a lot of interpretation. If you get into a meeting with 20 people, there's 20 different definitions and we have to start off by sorting all that out in terms of how we're communicating. As you add in this notion there and of this kind of contextual understanding that it has a representation, I am curious before we move on. Do you, and you've mentioned now robotics, do you see it as the same thing or is it kind of yet another variation of a world model using the word and stuff? How closely would you believe those two to be related as you're talking about that?
Just because both are big topics. Sorry, can I just ask for clarification? You asked my question, are models like our world model or a? Well, when you say world model, I'm just trying to clarify the same thing I did when there's 20 people in the room and you're asking what they mean by it. And as you mentioned, robotics and stuff, we all need this representation of the world out there so that we can do better about acknowledging the context of what we're working on, whether it's robotics or in visual intelligence, presumably. Are they, I'm just curious, in your mind, do you think
that they are very closely related or they kind of distinct ideas of what a world model is? What is your take on that? I would say that they're pretty related. I mean, to me, it's like, fundamentally, under the surface, what we're doing is we're trying to teach the model as much as we can about the world that we exist in and then asking for it to utilize that in some way. And up to this point, it's basically been mostly to creative mediums. But like, that representation that we're
talking about that, you know, and again, I'm trying to avoid this term because we, I don't know, it's a little overused, but the term world model, but I mean, this is what it's say, it is modeling, it's creating a representation of the world we live in, and then it is using essentially that intelligence that has to be able to act in it. So the idea is that it understands enough of these relationships and the actual physical properties of it. It's a really good foundation to build and train robotics on top of, gotcha. I hope that answers that. Thank you for putting up with that.
I was just curious. It's not unusual to navigate that. So go ahead, Daniel. Sorry about that. Yeah, from the non robotics person in the room, that was really interesting. I appreciate you going
into that. And I'm wondering, there's that kind of outworking of some of this where you are now kind of understanding the context that's in this model, maybe how it's representing the world. There's also things that I've seen just crossing my paths, whether it be kind of an e-commerce or, like I say, my phone, my text messaging. So on the web where these models are being more and more integrated into workflows, whether that be kind of like try on these clothes or glasses or whatever, or it's creative tools and like visual editing platforms, are you seeing that with your all models in terms of kind of what's the state of, and I know, I want to get into your model families here in a second, and they do different things, right? And many of them, I know there's a big, some that are open weight, so you might not know all the ways that they're being used, right? But from at least those partners that you're working with, in terms of today, what are some
of those creative, and maybe more practical uses of these models that you see out there, beyond just kind of the fun image generation kind of side of things? Yeah, yeah, absolutely. I would say, again, I'll come back to the moment we started to get context into the model that wasn't just text made a huge, like it was a fundamental change in what these models could do and how people actually worked with them. Our first model flex context, only could take one image reference it was mostly used as an editing model, but you could reference a, you know, a picture of a product and then tell it to generate like a nice product that is very nice. Then moving on, we introduced the Flux 2 family and then the Klein-Seedier smaller variant that now could take many references. And then now we're looking at, I mean, people,
and this kind of comes back again to what I was just saying of having this representation of how things relate and that the better we actually build this representation, the more interesting ways you can essentially tie in all of these different things. So I mean, you know, one of the most commonly used, or I don't want to say commonly used, but like kind of all these cases was okay, like clothing drive. Like this is a very, you know, here's a picture of me, here's a picture of some clothes. Beauty, please, you know, show me what this looks like. I've seen people like, I actually did some home decoration earlier this year just trying to see like what different couches and furniture look like in my place. One of the most interesting uses I'll say that kind of stood out to me that
I saw at a hackathon was, and I don't know how much that's going to be used for like a real planning purposes, but I just thought it was interesting is someone was taking pictures of fire exits in a building and then generating what it would look like in a crowd was trying to leave through the fire exit in an emergency. So they got actually like brilliant. Yeah. So they could actually gauge like what would this look under like emergency? Like where is the crowd like? And obviously there's,
you know, there's a generating component. So you have to take a little bit of grain of salt, but you still can get a general idea of this is what this would look like in this area. Without freerailing us, you just sparked a whole bunch of ideas on that in my head. So like, I'm just like, Oh, this is great stuff. Keep going. Sorry. If you've been listening to the show over the past few months,
you realize just how transformative agentic AI is, whether that's Claude Code or Hermes agent or custom built software that you're deploying for operational efficiencies or as new products to your customers, regardless of your maturity. Now, this is the world that we're headed towards this agentic AI world. And there's a lot of security and governance teams that aren't letting these agents go into production because of risks related to agency and autonomy. And how do you take care of things like prompt injections or insecure tool usage? There's a lot to take care of. And that's
why I'm personally spending my time outside of the show working with an amazing team of AI engineers to build prediction guard prediction guard is an AI control plane that you run in your own infrastructure behind your firewall. Developers can build on top of this control plane using everything that they want to use open AI and anthropic compatible API's MCP servers frameworks like Langchain. But all of this is plugged into a built in governance harness that enforces your organization's AI policies. And all of that telemetry goes back to your monitoring and alerting systems. I would encourage you to check out what we're doing at prediction guard.com slash practical AI.
You can schedule a demo with me and the team. And I'd love to get your feedback on what we're doing. So visit us at prediction guard.com slash practical AI. That's prediction guard.com slash practical AI. These are all so you mentioned the the flux family of models. So that's Black Forest family of models
or at least some of the models that you worked on. Could you just kind of give us a concrete that you mentioned a couple by name, but like give us a tour of the family of models and then maybe where I know that there's someone hugging face people can find them people can look at them but maybe just give us a little bit of a tour of the model family and then anything that you want to share about some of the some of the distinctions between them different ones of them. Yeah, I mean, we're I would say fundamentally like as a core we're still like a also we're still like a research lab that wants to push the boundaries. So with each one of our model releases, we want to try to like level up some capability of the model, not just you know, make a general improvement, but like really see some new versioning with it. So I mean, I can take back through not that many models. I take back through a little bit of history,
you know, it was a little I was around two years ago. Now we released the first our first family models, which was the flux series. And this was the first set of models that we created after we formed the company after a very fun but intense, I think it was around four or five months sprint to really like build our jobs here and get something out. And with that, we came up with this at the time, this essentially distinction between the models where we released the flux one. So we had the flux one pro model, which was on our API. We had the flux one dev model, which was this commercially
licenseable model, but the weights were open for people out in the world to use. And then we had the flux on Chanel model, which was a high speed like step to still model that I was totally, I don't know if it was MIT or catchy, but it was it was open to use for whatever purposes you want. And then from there, we worked on our flux to flux tools series of models where we realized like we needed a lot more control with the models. And this is where alongside this work started the context project to, okay, these people want a lot more control with the models, these models are learning a lot of relationships out we leverages, which led up to the flux context release, which I talked about this was our big edit release. Everything I mentioned up to this point was still kind of our like historical models. Now we're getting into our more up to date models, although they're getting
a little bit older now, but we have a something I'm excited coming. I don't want to debate soon, hopefully later this summer, that'll be very exciting to talk about when it comes out. But looking forward to it. Yeah, yeah, I'd love to get your back going. Definitely happy to have it to come back. But then we got
to our flux to series of models and flux to was a very big upgrade for us where we really push the capabilities not just on TTY, but also on editing, not just on single image, but this is where we also do introduce like the multi image that kind of on the edit where you put many people, many different items have all sorts of relationships. This is where we started to see, you know, not just these kind of more standard advertising use cases, which still were like the most common. We really want to support this, but like some more interesting how people kind of build, you know, like things like this fire exit thing or other types of stuff. And then right after that, we came to our client series of models, which was essentially like a size distillation where we really wanted to pack as much performance as we could into a small model for this, you know, for both use on our API, but also just for open weight release, you know, that there's many people out in the world who use our models locally and we really want to make sure they had something powerful they could use. So it was a text image and an editing model. It was a very fast model and it was pretty small in comparison to, you know, it was
even smaller than our our flux one series. And then we've done further, further speed up on that with our client KV model, which introduced, I believe for the first time, KV caching, which is I'll just say it's an optimization technique that's very common with the language model world that we were able to bring into the editing world to get a very, very big seat up on local editing. So people who wanted to actually like use these models locally, but also, you know, we also serve this as well. You know, now, now, since then we've done a couple blog releases, one fairly fun research, you know, research blog on some self-flow, which I'm repping partially because I am on there, not the lead author, lead author is incredible, he was, she's amazing. But, and then now we're kind of all pushing forward to our next big release, which is hopefully going to be, I'm quite excited about it. But that's
awesome. And just to follow up on that, I know some people out there are listeners are always trying things, you know, on their laptop or wherever they're pulling down things. I know for quite some time, when I tried to access some of these models and run them myself, it was either very difficult to find, find, you know, with my limited resources, the right, the right kind of configuration to run this, but also it was sometimes incredibly slow. So what to just give a sense of some of those, you know, more hardware optimized models that you mentioned, I think the client and other models, how maybe my question is, can I reasonably run one of these models on my laptop now and create a great image? Like, what's required here? Certainly, I'm not going to serve my production web app in
an enterprise environment off of my laptop, but just maybe give us a little bit of sense. Does that has changed on the LLM side, right, where that has continually updated and obviously the smaller models are not up to the same output quality as the larger models, but you can run a small model, you know, even on a CPU, now, at least for some tasks, that is pretty reasonable. So how has that progression happened on the visual intelligence or image generation side? Yeah, I would say I'll first say that the LLM world definitely has a lot more people working on it, so they definitely have a lot more of this up and down. That said, with our client series, I haven't run this personally, maybe not the best person in the world to speak on this, but I'm 99% certain you could run this on say like a modern M series, Mac, like a Mac with Pro. As for speed,
I don't, I don't have any numbers, I can quote because I haven't asked this myself, but this is always like a big kind of trade off in this world of how, you know, even on the language side, you know, you see the scaling of a, you know, like a lot of the new big open releases that people are excited about are getting into the hundreds of millions of parameters. And this is why we, you know, this is why we did like the client series, for example, because Flux 2 was a 32D model, it was quite chunky, you could run it on like a, you know, higher end local computer, but it was slower to run locally, you needed some more powerful, you know, computing, this is where like the client series popped in, but it's always a trade off because we want to push the, you know, we want to push performance and make something that we're really proud of, and then we want to try to bring that again back into a new, a smaller faster scale, and this is something we've continually done, you know, with our Flux One series, we had Chanel, we had Klein, I don't want to promise anything going forward, but we definitely want to like keep bringing this like bigger power that we generate into these into these smaller models that people can hopefully run locally.
Absolutely. Super cool. I'm excited to try some of those with like, as we are starting to wind up, and you guys have done such some really, really cool work in the space. One of the things we always like to ask is we're closing out these episodes are kind of like we want to get a glimpse into your head into your thinking about what's to come in, and kind of to frame it a little bit as you're thinking about like, you know, you're, you're out of the day's work and you're, you know, enjoying your evening and your mind wanders and you're kind of thinking about like down the road, what, you know, what you want, what your passion is, is taking you and what you'd like to see. Can you share, can you share a little bit about what the future that you would like to, to create with visual intelligence in terms of what's next that you guys haven't addressed. I'm not asking for a product
release so much because I know those, those need to stay in, but kind of just like, what's, what, you know, what is your aspiration, what's your dream in terms of where these kind of capabilities might ultimately lead and love to love to get your, your, your kind of, your dream context, if you will, as we finish up. Sure. If I'm allowed today to give two split answers. Absolutely. Yeah. There's,
there's a totally fine. There's kind of two areas that constantly sit on the front of my mind. At some point they'll hopefully become one, but I imagine the next, and this is no, no glimpse into saying what we're doing, but I just, this is so I'm personally excited about fair enough. Well, I would say the two areas that I'm the most excited about is one is long context truly multiple multi-modal models. So now nowadays, obviously, we're seeing lots of people work with agents constantly. And we're, you know, on our front where we're having these, these more like
continuous generative models, we're starting to introduce more and more, you know, contextual usage with these like references. And, and I'm looking for, I'm looking forward to when this kind of bridges a little bit more. And we have models that not just can like, okay, you have, say, the agent calls the generative model, but when this kind of becomes more or less a model that not just can, you know, do work in the text and language and agent, but actually maybe can think visually as well and generate audio for you and has the full context of say all the stuff you've done over the last few weeks, you know, you don't need to say give it the reference, give it the right prompt. It just like already has this context and can references is needed in this like continuously. This excites me a lot. The other side I'll say is the real time stuff that's, you know, we're seeing a little bit of it now
there. I think that this is like very early. And I think there's going to be very exciting for real time video audio to flex interactions, you know, being able to, I don't know, you see a little bit of these like interactive, you know, stuff like genie or play the game, but also on the side of, you know, you can bridge this back into like robotics or it needs to take in the real world in real time. I think that's it. Yeah, I was going to say that sounds really familiar in terms of interest on that side of things. So yeah, really cool. Great conversation. And you're leading to what,
to what Black Forest Labs is doing was also really good contextually in terms of kind of explaining. So hope our audience got a lot out of that. Dustin, thank you very, very much for coming on the show. Great conversation. And as you, as you hinted, there are things to come and looking forward to having our next conversation as things move forward a little bit. Absolutely. Thank you so
much for having me. All right. That's our show for this week. If you haven't checked out our website, head to practical AI dot FM and be sure to connect with us on LinkedIn X or blue sky. You'll see us posting insights related to the latest AI developments. And we would love for you to join the conversation.
Thanks to our partner prediction guard for providing operational support for the show. Check them out at prediction guard.com. Also, thanks to break master cylinder for the beats and to you for listening. That's all for now. But you'll hear from us again next week.