Step-by-Step: The State of AI Filmmaking Workflows

Joey and Addy break down different workflow of creating AI-generated films, exploring which tools to use at each stage and how to maintain consistency. They examine text-to-image creation, character development techniques, and how video-to-video models are shifting the entire approach. Plus, practical tips on using Runway References, ChatGPT, Flux Kontext, and more AI tools that are reshaping production pipelines.

--

The views and opinions expressed in this podcast are the personal views of the hosts and do not necessarily reflect the views or positions of their respective employers or organizations. This show is independently produced by VP Land without the use of any outside company resources, confidential information, or affiliations.

Transcript

0:00

The top three tools that are just, if you're kind of just trying to like in a chat mode, easy to use, that kind of hits the bases is gonna be Runway References. Yeah. ChatGPT and Flux Kontext. Alright, welcome back everyone to Denoised, we're gonna do sort of another deep dive episode, uh, today looking at. AI filmmaking workflows and what sort of the typical workflow that a lot of people figured out is happening. Uh, tools that are used along each step of the way, and also how that's changing a bit with, uh, more impressive video to video models. Yeah. And if you missed our last episode, we did a deep dive on ComfyUI and some of the fundamentals of image generation and image workflows. Uh, give that one a go. Yeah. Before this one if, if you like. Yeah. Yeah. Check that one out. That goes into the nitty gritty of like how AI image generation works and this one we're gonna be a little bit more inflow of more creator driven. Yeah. And just sort of the workflow and different tools. And most of these are gonna be off the quote, off the shelf tools. Yeah. Uh, uh, websites that are pretty, uh, consumer friendly to use. Um, but sort of how you have to kind of think about the process mm-hmm. Each step of the way. And so a lot of this is adapted well, first off, I started building out this flowchart because I started getting asked questions. You know, people are like. I want to, you know, make an AI movie, or I wanna use AI to help make a film. What do I do? Where do I start? Where do I start? And then I was, I was thinking about it. I'm like, well, you know, I'm like, a bunch of different tools and options would pop into my head. And it's like, there are a lot of questions of like. How you wanna think about this. And also there are multiple tools along the way that you sort of have to use ish. Tools are getting better at kind of integrating to kind of be a one stop shop. Yeah. Uh, so you don't have to hop around between a bunch of tools, but we'll cover all the tools just so you can,'cause some are better at some things. Yeah. And some are better at different things. And it depends too of like what you're. Need. It's like, do you need a more action based shot? Do you need more static shot? Do you have characters performing? Characters talking? Yeah. Multiple characters. Uh, so there are all kind of things you gotta about. Yeah. And regardless of what tools you use, you, uh, we have a very agnostic workflow here that you'll find beneficial just in general. Mm-hmm. As the tools change and evolve. The flow chart here that we share won't really change. Yeah. I mean I've seen, uh, this also evolved from watching all the creators on Cinema Synthetica Yeah. The AI on the Lot filmmaking, competition, and they all basically use some form of this and as hard as like text image, image video. Yeah. That's sort of the gist of it. And that was sort of in some fashion Yeah. The workflow that all of them use today. Today also, that is still a good workflow too, if you just have a computer. Mm-hmm. You don't have any resources, you don't have a camera, you don't have like. Access to film some stuff. Mm-hmm. You can make something visual. A film. Oh yeah. Uh, just using your computer with that workflow, you can make a pretty convincing film if you do it right. Yeah. And have a good story. And this all goes to the root of, you know, good story and, and good. And something to Yeah, to to tell. Amazing visual and filmmaking techniques. Amazing visuals won't save you from a bad story. Yeah. E. Exactly. Exactly. Alright, so first up, starting out with you gotta have a good story. Yeah. And assuming you've got a good story, you know what you wanna say, you wanna shot design that story. You could do this yourself and obviously each step of the way you could kind of figure this out yourself, or you could lean into some of the AI models. Mm-hmm. Um, LLM. So in here, in this case, for the tech stuff, the general LLMs are gonna come into play. So either, uh, Google, Gemini, mm-hmm. Uh, Claude or ChatGPT. I, Gemini 2.5. I've been hearing a lot of really good stuff and I've been using it a little bit more and more. Mm-hmm. Uh, especially with like shot design or kind of creating prompts for shots. I use cloud a lot. I found cloud's always been sort of the best at Okay. Creative writing tasks and stuff like this. So you could start off with. Is that, uh, Higgsfield specifically designed for stuff like this? Uh, Higgsfield would be more later on. Okay. Of like, you want, once you're like, uh, you wanna start generating images or Gen or turning images into video. Got it. So this is still in the kind of ideation, ideation planning phase. Yeah. Like, so first off, you could figure out, okay, I wanna make an action scene of a, like a guy, you know, being. Chase through a, a mine tunnel on a mine car temple run Again, what kind of shots would I need? You can have help brainstorm the shots. Yeah. And then from there you can also help generate the prompt. Yeah. To generate your, your image. So go from text to image. Before that though, you probably wanna figure out a consistent style or visual of how you want all the shots to look. Uh, so if you don't, so. You don't generate a new shot each time and it has a completely different visual look. Correct. A couple ways you can do that. One is you just help you describe what kind of look you're going for and you create a descriptive text prompt. Yeah. Uh, of the style. Right. Other way we should talk about Comfy is you can make a LoRA way better. Way better. Yeah. And you can guide it with an existing image Uhhuh. So if you have like a cyberpunk 2077 type look, you know, futuristic Blade Runner risk thing, then give the model imagery of that city. Mm-hmm. And it'll sort of start to get a good look and feel of that city from just a few models. From a few images. Would you use existing images of just like reference images from other movies that. Have the style you're looking for, or would you try to generate, like in Midjourney images that have the style and then train it on those images? I don't know. I mean, there's an ethical question there. Uh, I would just, uh, generate stuff specifically for influencing generation. If that makes sense. Right. Create an image. Like let's say if I create text image in stable diffusion. Yeah. In that style, 20 images or so with the cyberpunk style and feed that to train the LoRA. Yeah. Okay. Other option, other way for creating style, if you're working in Midjourney maternity has SREs, which are these codes that trigger and call up a specific style. So Cool. Yeah. And so, and you could also. Similar to the LoRA, you can give it a bunch of images and it'll create a style. Yeah. That'll, uh, call it up by that. This only works if you're working in Midjourney, but it works really well. Yeah. And, um, the technical term is style transfer, so it'll sort of take, uh, whatever, uh, the image is generating and then, uh, apply like a, a filter mm-hmm. On top of it. A style transfer filter. Yeah. And then the other way is you can use reference images. And so some of the tools like Runway References mm-hmm. ChatGPT uh, you can or yeah, Google image. Uh, flux context. Flux context. You can give it another image and say you want the style of this image. Yeah. And not necessarily any of the details, but we just want the aesthetic look of that image. And you may have to run it a few times, but you can use that as a consistent to try. That's how we got to all of the, uh, Spirited Away Miyazaki stuff. That those were all style transfers also. Right. But that was, uh. It was, it knew that style. And those were just people calling up that style. Exactly. It was, they weren't giving it an image of like from Spirited Away and then giving it an image. ChatGPT already did that for you. Chad knew what that style was. But let's say you're trying to create a unique original style. Yeah. You can give it an image. Yeah. And your image of what you're trying to generate and, and try to have it, uh, match that style. Yeah. Like a, like a Wes Anderson film has a very specific look except original style. Right. We want you to make original stuff. Yes. I was just pulling an example. But yes, that is a very distinctive style. Okay. So you wanna kind of, so this helps because shot to shot you want to. Make sure each new first frame feels consistent. Like it's in a consistent world with a consistent cinema look. Yeah. And this is where you as a creator really matter, right? Yeah. Like storyboards is what's gonna drive everything down the pipe. And so if you have an eye for blocking eye for com, frame composition for lighting, for cinematography, this is when all of that's gonna come into play. Mm-hmm. And then the other thing that's gonna be an issue with consistency is characters and making sure that your characters look the same. Yeah. Shot to shot. And so ways to do that. Similar to the style, you could either, if it's sort of a generic e character, you can just have a very descriptive text prompt. Mm-hmm. Um, this is sort of what a lot of people do are currently doing with Veo 3. Mm-hmm. Because Veo 3 is currently just text you video, there is no, you can't give it a frame. So in some of the launch videos, it was like the SWAT police Yeah. Action video that Dave Clark did. But that's kind of a easy one to do because yeah, the faces are kind of covered. We're not really seeing the details of their faces, and, and it's something that's sort of easy to prompt and get a very consistent looking. Physical character. Right. But if you're, you know, actually building characters where we see their face and they're talking and performing, LoRA's is another option as well. Yeah. You could turn that onto character. Exactly. And, uh, you know, as human beings, I, I bash this with a hammer. Every episode is like, we're so susceptible to, uh, looking at people's faces. So the minute that you have a slightly uncanny valley type face, a face that doesn't look quite natural, like instantly a project breaks. So you want to spend a lot of. Yeah. Time and effort here. Yeah. Okay. And then we'll talk about actually making them perform and talk later, but that's, yeah. The other issue. And then also, you know, going back to the tools that we've talked about where you can just sort of give it a single shot. Yeah. You, um, a lot of other tools, if you just have a photo of a person that can sometimes work mm-hmm. And you like Runway, you just give it a photo of the character and, you know, say, oh, I want them in this scene. Hopefully sometimes it'll like. Recreate the person pretty accurately. Sure, sure. Another trick that you could try to do is to create a character sheet. So a single image Yes. That would be much more of like your front face. Yeah. Your side profiles full body. So like a character reference sheet. Yeah. That's a single image that you can give to these models and that that is done in animation all day long. Right? Mm-hmm. When they're in character development, uh, they're, they, they call 'em post sheets mm-hmm. Where all that is created in CG traditionally. Mm-hmm. But now you're doing it. In Photoshop and then putting that up onto your control net model. Mm-hmm. Uh, we, Mickmumpitz does a good job. He does, and he has a whole comfy workflow where you can give it a single image of character and then it'll create an entire post sheet. Yeah. Ideally, if you have, uh, 10 characters in your film, every of those 10 characters needs their own. Based on, and then they'll need their own post sheet. Mm-hmm. Even with alo, you'd want to put a post sheet. A post sheet would be like, uh, let's say you have a hero male character, so the post sheet would be male standing. Like this, standing in the back side, you know, threequarter view, face, closeup, face smiling. But I guess I'm saying, would you train the LoRA on that pose sheet or where does the two, you have the character, LoRA, and you have the pose sheet. So how do the two come into play if you need both? The LoRA build happens before the pose sheet. Okay. Right. The, you use the LoRA to create the post sheet, and then the post sheet influences the generation because you're the character's moving in space. Okay. I see. So it'll use both the LoRA and the POI together to figure out what that human should look like at whatever angle or whatever framing. Okay. Got what you're saying? Mm-hmm. Okay. So now we've got, so these are sort of the foundational things you kinda wanna figure out before going and to start making your first frames of what is your character, all your characters gonna look like, what's your style gonna look like? Yeah. And also, I didn't get it, but like you're, if you're having a, a location, that's a big deal. Yeah. Yeah. The scene, the, the, the location. Um, one trick that I was. Did with the AWS, the Amazon film? The, oh, no, no, I'm talking about that, that Culver Cup filmmaking comp thing that I Oh, okay, okay. That competed in, I, I, it was, um, I had a, like a 3D room in Unreal, but then ex you could export a 360 panorama. Oh, interesting. And then I style transfer the entire panorama, but in that way, I could just use portions of the background. Oh, I love that. So that if it was like a conversation like this, it was like, oh, if I had the reverse shot, yeah. I would at least have a consistent. Background scene. Yeah. And, uh, spatial consistency in image generation is still a, for like frontier thing. Yeah. Uh, I know Flux context has a decent way of approaching that and so does ChatGPT, but again mm-hmm. You are, you are essentially short cutting that process by using a 3D model of a world and giving it Yeah.'cause they have like a world understanding. Yeah, exactly. Yeah. Yeah. I mean, we're gonna keep going back to it, but I think the three, the top three tools that are like, just, if you're kind of just trying to like in a chat mode. Easy to use, uh, that kind of hits the bases is gonna be Runway references. Yep. ChatGPT. Yep. And flux context. Yep. Those are sort of the top three of mm-hmm. Of just being like, here's an image of a thing in a world. And, and that's understanding's sort of state-of-the-art today, like, uh, recording this. Yeah. And that is a only for this, uh, pre-production portion of the production that we're in. And now we take that and everything we've doing, and now we start generating the first frames Yes. For each of the shots that we want. Yep. And so. This is either gonna be text to image, merging it with if we have LoRA's or if we have reference images Right. Uh, in our chat interface with, uh, some of these tools. And then also this is, we're also using the. Any of the LLMs comes, helps come into play to help you refine and write out structure, prompting, structure the prompt, help you write out the prompt, at least give you, um, it's also good to just have it write the prompt as a starting point and you can kind of come in and tweak it, but mm-hmm. They'll write much more detailed prompts than I'll think of off the top of my head. Yeah. There you just modify the words some, um. VMs, visual language models, um, out there. I haven't done too much research into this, but you can give your image to A VLM and then it'll caption it for you correctly and give you a suggested prompt and stuff. Oh, like, so if you like, uh, have a reference frame, you're like, I want this kind of shot. Give it that frame, turn it into a text prompt, and then use that as a, use that as a input to your, I think Gemini's pretty good at that.'cause Gemini is multimodal and you can give it input mm-hmm. Images and videos and it'll describe or tell you what's happening. Yeah. And remember what we covered in the last video, the clip encoding? Yeah. It's, it's bidirectional. So you can go from text to image, image to text back and forth pretty easily. And that's what VMs do. Okay. Yeah. Okay. So yeah, that's another good, uh, trick too, if you like, have a reference. Shot where you're like, I want replicate this. Yep. Type of angle, feed it, turn it into a text prompt, and then generate your first frame. Okay. So also just to round up as sort of the common tools for generating the first frame. This is gonna be all of your text and image, uh, generator. So Midjourney Runway, Ideogram, Flux, ChatGPT Imagen. Imagen. Yep. Reference. You could use it with a references feature, which we talked about also. The fourth one is Imogen as well. Google's Imogen, which does have pretty good. You can give a reference, you can give it a character Ference. Nice. A style reference. Yeah. Location. And it does a pretty good job of having that chat conversational workflow to help modify as an AI creator that that is the key. Right. Otherwise it's just a random thing every time. Yeah. The more control you can have, the better. That's the thing you're gonna be fighting all the time. And then we've talked about this before too, but you could also kind of do a hybrid approach with Blender or Unreal. Yeah. And very sort of kitbash scene together. Yeah. But at least. Roughly frame up a shot and use that as a base image to style transfer? Yeah. If, if you're pretty handy with Unreal engine, I mean, use Metahumans, pose them, right, and then use a, a camera to kind of block the characters and then bring that. Still image into one of these things as a reference. And then there might be a final stage too. Like sometimes you might get an image, but maybe it's like 90% of the way there of what you want. Yeah. But it still has some issues, but it's like not worth it to keep regenerating it. So then you might take the image and you have to, might have to modify it either with, in painting, in painting to like, uh, change specific aspects of it. And most of the image tools have some form of inpainting. So like Midjourney Ideogram? Yeah. Explain to the viewers. Yeah. So painting is like, okay, we wanted to keep this image. Is, but like I want to change just this computer to regional correction. Yeah. I just want to, I'm keep everything the same, but I wanna change my coffee cup to like a big chalice. Yeah. Or something. And so then you would take the image and you would like paint over the coffee cup and then you would text prompt to make it a chalice. Yeah. And it would generate the chalice, but blend it theoretically. Hopefully blend it in with the image. Oh yeah. It doesn't really, I mean, most inpainting models are really good. Photoshop has Firefly built in and Right. That is one thing I'm just pretty good at. Yeah. Of taking, you know, and also, I mean, sometimes in some cases you might just straight up need traditional Photoshop. Yeah. Just modify. Exactly. Yeah. And, and don't, don't be shy to go back and forth between traditional tools and AI and back. Yeah. Yeah. It doesn't have to be AI all the way. So like a lot of times, uh, this stuff still needs a lot of massaging. Yeah. And a lot of times even just traditional compositing or trying to think of building out your elements separately. Yeah. Might be the better option. We won't get into that here.'cause that gets a little bit more Yeah. Complicated. And this is. Kind of more of a vibe, filmmaking flow we're talking about. Yeah. I, I find that with multiple characters, it's probably easier for you to generate a consistent character and then composite them in Photoshop and then take that reference frame as your image generation, uh, video generation frame. So generate the characters. As separate elements and build it all as one frame inside Photoshop. Yeah. Yeah. Like it doesn't have to be exactly, you know, uh, geometrically correct and stuff like that. Uhhuh, the generation will take care of it, but at least you'll get, you know, your three hero characters in the same shot on that reference frame. If you try to just generate that in ComfyUI, you're looking at a pretty crazy workflow. Like three different LoRAs, three different character sheets. Yeah. So on. I mean, I have seen some hacks now with Flux Kontext out of like combining. Images into a single image and giving that to flux context and that working pretty well in Comfy. You've you've showed that in one of the episodes, I think. Yeah, yeah, yeah. Uh, right. It was like a Yeah, it was a hack or something. Yeah. Another note. You get five people in there. Yeah. Yeah. And then also, I mean, yeah, I've seen, uh, Runway demos where you could just kind of, I think they built in a sketch tool in the Runway because people are doing this so much, but you could just kind of roughly sketch out Yeah. How you want the scene blocked and it sometimes does a pretty good understanding of, sure. Replicating that in the image. Yeah. Okay. And so then you've got your first frame. Mm-hmm. Then you wanna turn that into a video. Yeah. And now there are some, most of the tools will support Image two, video, first Frame Image, and some tools have more control where you could give it both first frame and last frame, or first middle, last frame, but the Gen, the general broad overview of like all the tools that will support Image two. Video got your Runway, Veo 3, uh, Midjourney just launched. Video and that's been pretty impressive. So yeah. People are impressed by it. Yeah. Kling, uh, Hailuo the Chinese models. Yep. I talked about that also, that, uh, get really good reviews, uh, Luma ai. Yeah. You can also run this on your own. So Wan 2.1. Yes. Thank you. Which is by Alibaba. Is, is a very popular ComfyUI, video generator model. Mm-hmm. Oh, Seedance. Seedance That was a new one too. We just, yes. Yeah. From uh, ByteDance. Yeah. If you have like a RTX 3090, 4090, like last year's GPUs, I would say like a 720p video for 30 seconds, you're looking at like an hour. Mm-hmm. Or so, like it's not that terrible. It's like a old school CG render timelines. And then, I mean, this part is always, still feels a bit weird to me 'cause it's like you do all this work to get the. Image the first frame, but then it's just like, well, I'll just give the first frame to the model and like go for it. Let it do its thing. Yeah. So some of these you'll, some of the models, you'll be able to give it a text prompt to try to describe what you're looking for. Yes. And some of them are, a lot of 'em now have sort of prebuilt camera moves to, which can also help get what you want. Yeah. You wanna resolve ambiguity as much as possible. Yeah. Runway and Kling and their older models, they had a kind of a camera control. Movement where you could sort of map out how you want the camera to move and Runway. That was only in Gen-3. It's not in Gen-4 yet. Okay. So Gen-4 doesn't have as many features as Gen-3 does, but I'm sure yeah, eventually that will get added. Uh, we, we didn't mention, um, the one you mentioned Higgsfield. Yeah. Higgsfield is specifically built for cinematic shots. Quote. Yeah. I would say it's a bit more for like advertising or like. Very dynamic, high motion shots. Yeah. For like, uh, an advertising spot. Sure. But try them all. Yeah. Yeah. I mean, a lot of these is just like, yeah. Whatever tool gets you the shot. The, the pro here is all these tools are really easy to use Uhhuh, and they're relatively cheap, right? Yeah. If you're looking at like, how many cents, how many dollars per video, it's not that much. It's not like gonna break your bank. The con here is the control is still something that we're all trying to figure out. Yeah. Yeah. Getting the control of what you want. Um, yeah. Which there are some good work, uh, examples a little bit later, but other thing that might come into play too, if you have characters mm-hmm. And you need them to talk, then now you gotta introduce another tool. Yes. Uh, so that could be something like, uh, uh, Hedra, uh Yeah. Which has really good character performance. Um, you can give it an image Yeah. Of the first frame and give a direction and the audio. Maybe HeyGen Hagen's another one. Yeah. Yeah. Runway Act-One. Yeah. Which also has not been updated for Gen-4. So it's still Gen-3. Okay. And then there might be elements of compositing. Yeah. These things, because it's like you might have elements of like where the character talks and you need the lips to move, but then also you need camera movement. So you might, this might come into play where you have to start compositing. Yeah. Or treating like the performance separate from the camera move. Yeah. And combine things. I would say I would vo VO three probably does the best job of animating faces for speech and matching speech. Yeah. But Veo 3 is only text. So limited. Yeah. Video, so you're gonna have to have a very descriptive prompt. That's why it worked really well in that commercial. The PGA did the Liquid Death one, not the liquid death, the, it was a gambling Kalshi, Kalshi, Kalshi, something like that. Yeah. But it was like, that was, that was a good one. It was good, but it was like every shot was a single. Scene shot with their own characters in that shot. So you didn't have to worry about a character looking the same shot to shot. And there wasn't dialogue, right? There was no dialogue in it. No, there's dialogue. It was the, it was the news interviewer like kind of interviewing people. Yes. But every shot was like a different. Person, different interviewer. So, uh, it didn't, yeah, it didn't matter. And then that's why you see a lot of sci-fi stuff with like, masks on no faces. Yeah. Because it's still hard. Yeah. Easier to, to, to deal with the performance. But stuff's getting better. I mean, Hedra uh, Hedra's new version is really good. We're testing that with a project, so We'll Yeah. Report back. Yeah. So just another thing to think about, you know, I mean, Joey, what about this like. Can the human performances be actual humans with a real camera and then you cut to your AI generated shot. Right? And that's the thing too. You'd have a hybrid workflow, right, of real people in front of a green screen or the Totally, uh, using another workflow like Lightcraft jet set and then you, you new composite that humans face onto your generation. It's, oh, I see what you're saying. Actually, just do the face onto the generation. Yeah. And that's interesting. Mm-hmm. Then it could, yeah. If you're good enough at Nuke. I was gonna say, how complicated is that? A Nuke, if you're, well, you gotta match, move the camera, right. And then take that match, move camera into your generation, so the generation's from the same angle. And, uh, if it's a static shot, it's fairly easy, but if it's moving, then you're warping that face. Yeah. Yeah. I was thinking of more just like if you shoot a person mm-hmm. In front of a green screen. Yeah. With something like Lightcraft, with like a, with a. Lofi 3D background. Right? And then you take the background separate and process the background Exactly. Separately. That totally works too. Could, and then run it through, uh, be Wolf for relighting, so the whole scene matches. That's another workflow that a lot of people are exploring. Yeah, I think the AI faces and AI speech is still. Not, not there yet. Yeah. I mean we've seen AI is very good at these like one shot action shot generations, but having that consistency shot shot, imagine a 90 minute film with, there's like hundreds of lines of dialogue. Yeah, right. You want a real human, yeah. Yeah. It gets tricky there. Yep. Yeah, so that's high level overview gist. That's why there are a lot of steps and tools. Uh, there are tools that are sort of combining this all, so you can kind of do it in one stop. Yep. So like Freepik has a lot like kind of one stop shop. Mm-hmm. Um, even Hera, a lot of 'em are building, connecting to the APIs mm-hmm. With other tools so you don't have to bounce around. Yeah. A lot, but. Another thing to think about that's sort of changing a lot is, uh, since Luma AI released the Restyle video feature. Yeah, that's great. That's been blowing up and that's been a kind of, kind of shift. Yeah. Like thinking about this completely, where uh, you can just shoot a very low-fi object blocking video. Uh, I mean, with Jon Fingers, his demos are crazy. Jon Fingers are so like the one where he's just like swimming on a carpet and, but it's an astronaut in space. Yeah. Or it's like of someone with the sheet moving and then it's like a stormy ship. Um, he he's really imaginative. Yeah. Or he had like two shoes, like pulling the strings of the shoes and it turned into like a race car scene. It was like a race car. Yeah. Yeah. So th this and I mean, yeah. You know, we, we are old. Like, I just can't imagine what, you know, some like kids who take this and run away with it 'cause it's like, it could just kind of re you can just rema you, you don't even have to have the same rules. Yeah. Of like matching a shot. Like if I think of a race car shot, I would still probably be like, oh, I need some toy cars and I'll turn the toy cars into race cars. Yeah, but, and it, it doesn't have to be video to video. Uh, think about this way, this will blow your mind. Okay. You can go from video to video to video to video. Like you can keep iterating on it in Luma to get the quality and this. Stuff that you need. Interesting. Yeah. And I'm sure Luma would like that.'cause that's like generation, so yeah. So this type of workflow I think could, could just totally, absolutely. Throw a lot of the stuff we just talked about out the window. Well, not necessarily because you still need the solid image references, but now you're shifting to a more of a video to video workflow. Yeah. It's like, oh, what if I just take my phone and I could just shoot with a few people or I could just shoot with some toys on a table and now we've got like an epic action film. But also I like, you know, when I'm like, oh, it's crazy that you just give a first frame. To the image and then you like let it do its thing. Yeah. But now you know, Luma is mimicking the actual movement that you do. Cinematography everything. Yeah. Luma does take a first frame. You can either not give it a first frame or you can give it a first frame. So you could combine the elements up to making that first frame, and then that camera control could be, you know, Luma or it could also be. A blender or an unreal engine. Yeah. Recorded camera movement. Exactly. You restyle the first frame. Mm-hmm. And then now you're really driving what the image should look like and how the camera should move Yeah. In that space. Yeah. The, the, the biggest benefits to physical production and principle photography is character performance. Mm-hmm. Direction from the director on the performance and then cinematography. Mm-hmm. Like you wanna preserve those aspects of filmmaking as much as possible. And I think Luma allows you to do that. And yeah, I mean, video to video has been pretty slim on who's offered it. I mean, Runway was the only one that did it for a while and it's still, they haven't updated it. It's still only for Gen-3. I mean, I'm sure it's only a matter of time before Gen-4 has video to video. Cristobal. Let's go. You wanna announce it on the podcast, Cristobal, Come on on. Yeah. So also, once that comes out too, then it's gonna really be, 'cause. Gen-4 itself right now is pretty crazy. Yeah, it is. So once, once very, the video video comes out. Yeah. That's gonna be another. And it's such a exciting time. We live in like all of these tools we're talking about. They're all great. Yeah. Not one of them is Buster. Especially in hra. Like if you look at the stuff that, uh, people posting things like from two years ago. Yeah. And it was like, you know, just abstract, bizarre world. Yeah. Like Vincent van Gogh's. Style transfer. That's all we got. Yeah. A few years ago. And so, like how fast this has come in just two years. I know. It's crazy know, and it's already pretty usable, but we're gonna get to a point where all this stuff is effortlessly usable. Yeah. Yeah. And there won't be a massive, um, flow chart workflow. Yeah. You won't need Joey and Addy anymore, but Hey man. We'll, we'll, we'll move to cooking show. You'll have AI generated Joey and Addy. Yeah. But I think I a good place to wrap it up. Yeah. So, uh, yeah, I mean this was. This is a lot of the tools and workflow that I've been seeing, but if you've got your own workflow or other kind of things that you've been using, uh, that we didn't cover, please let us know in the comments. I would love to hear it. Let us know if this is the type of content you wanna see. We'd love to make more of this stuff, but it's up to you, the viewer. So give us a like and give us a comment. The power is yours. All right. See you next time.