 Hello everybody, welcome back to another AI video. In this one, I'm going to show you a new free AI image generator or AI art generator. It's called Leonardo.ai. And the link is in the description below. And here's the thing. It's really, really good. It's similar to things like Mid Journey, and Blue Willow, and Stable Diffusion, Playground AI, et cetera. But this has a few features that the other guys don't have. And they might be worth a look here. Also, did I mention it was free? Yeah, that's a good thing. It might change in the future, but it's free right now. Anyways, here we go. Log in like I've done here. And this is what you're going to see. At the top, you're going to see featured models. And these aren't models in terms of people you're going to draw. These are actual AI models. So if I click on one of them, well, let me show you a few of them first off. So you'll see a few of the different styles. You'll see here, paper art style. This image is reflective of a sample image using that style. Illustration, character portraits, pixel art, isometric, et cetera, et cetera, et cetera. A lot of great styles here. You can go through them, of course, on your own time. But I just wanted you to know that you could see all of these styles. And they're kind of like templates. They're tuned models so that if you click on it and use that model, you're going to get stuff in that style. So anyways, I'm going to go ahead and use, I don't know, let's go with deliberate 1.1. Click on it, and then here we go. If I want to generate with this model, I just click this button here. If I want to know what it does, well, who created it, what base model uses stable diffusion, 1.5, et cetera, et cetera. And if I want to look at some examples done using this model, here's some really nice ones here, this attractive red head, 1950 style diner. All right, cool. Let's say I want to create something in this image style or using this model. Just click generate with this model. And we are on our way to using this for free. Okay, good. Now I'm going to go ahead and select the number of images on the left side. I want to, if I click a prompt, I want it to make four versions. So instead of the default one or two, I got four here on the right side here. You're sort of in the middle. Deliberate 1.1 is selected. If I want to use a different model, I just click the down button. You'll see Caravaggio. That's my own personal model that I made. Civ 6 leaders, which is kind of cool. Isometric fantasy, stable diffusion, custom model, et cetera, et cetera. But I'm going to go ahead and show you how to do it with this one here. And then you can go none or Leonardo style. Leonardo style is just sort of like, it gives it a little bit more freedom to do some, you know, some more creative stuff. So depending on how much you want to stick to the original model and then negative prompts, we're not going to do that. I'm just going to go Bulldogs playing poker and I'm going to click generate and I'm going to get this ball rolling. So while I do that here, you're going to see that it's creating four images all at once. It uses four tokens, but again, I've got 250 a day. So I don't know uses up uses up tokens. But again, this is free because you get so many tokens on the left side here. While it generates, you're going to see you can select different dimensions. You can go with the smaller dimensions, the bigger dimensions, the bigger, of course, take a little longer. You can adjust aspect ratio and width and height sizes. So if I clicked on this, for example, if I wanted this for a YouTube thumbnail, I would go 69. And if I wanted it to be a square, I'd go one to one, et cetera, et cetera. Guidance style and step count are just basically the complexity and the number and basically how close it needs to adhere to your prompt. So if you want to give it lots of latitude to go crazy and come up with some abstract ideas, just increase, decrease those counts. Also, last thing I will mention while I got you is you can upload an image to use as an input. So if you wanted to say, hey, I want Bulldogs playing poker, but I want it, you know, based off of this Zeus looking thing. So I want it to look kind of like Zeus has something to do with it. You can do that and then click Generate and it will use this as a input guide. I'm going to delete that, but I do want you to know about it. Anyways, here we go. Let's see what we got. Here are our four options that we got. And quite frankly, I kind of like them. I look at all these cute little Bulldogs playing poker. Awesome. Right. Let's see what else we got. OK, this one's good. You know, these are great right out of the box and we're not even upset. OK, this dog. These boys look like they're in some business. Oh, this one might be a little bit weird. But and then let's go with the last one. OK, this guy looks unhappy. He might have just lost. OK, so there you go. We're going to use this one here. And then when you hover over it, you get a few options. Download the image. You can unzoom the image. This is this is unique to Leonardo AI. You can go ahead and just remove the background. Think about Photoshop masking and removing or selecting subjects and masking. You can upscale the image, which I'm going to do reason why I want to do that. It takes a few seconds to do it, but it's very, very cool because a lot of times you're going to get an image like, let's say this one here and it looks great, but sometimes like the facial features just don't look right. Like they all kind of look kind of sad. And what happens is you upscale the image and then all of a sudden the the facial features or the backgrounds they become a lot more dialed in and cool looking. And then it actually looks like a good image. So I'm going to show you on this one here, the other unique thing here, which is let me see here, unzoom image. This basically acts like a camera. It just backs the camera up and puts more into the shot. So while that one works here, I have gone ahead and finished this one here. Click back on it. And then you're going to see at the bottom here, normally it says original image. And now you can select upscaled image and the upscaled image is a lot more detailed. Looks great. If I like this shot, I just click on the download image and here it is. I would just click right click, save image as and I'll just put it in my I don't know, put it in pictures and that's it. I've got the image. If I click on it, it's going to open Presto. How cool is that? It's just just awesome. So there you go. The other one was what I just did the what did I do that the unzoom image? Yes. So here is an example of an unzoom. So here's the other unique feature. Original image, unzoomed. When I unzoom, you get more of the dog. It looks like it's wearing a skirt or something. I don't know what the hell's going on there, but you just get more in the shot. A cool feature. I like it. All right. So we've gone through all of that. Now I want to show you what really makes this super cool, which is the ability to train your own image sets or create your own model. So I'm going to click on this back button here. We're going to go back to the front page. And before I start showing you how to do that, do keep in mind that again, you guys can look through the recent creations from the community. And then if you like something, you can actually just go ahead and take it. So let's say I like this, I don't know about furries. That's a little weird. I don't know what the hell that is, but I'm going to steer clear from that. There are some NSFW. So I just clicked on this image here. And when I do that, I can go ahead and remix it. How cool is that? I can copy the prompt. I can image to image using this as an input into an image I create. You can download it as is. You can share it. You can use the same model. So if I want to create something similar to this, I just click this button here. And it just generates with this RPG 4.0 model. Like just so much cool stuff going on here. And again, I can just download it and take it, right? So I mean, just so cool. So yeah, just a heads up that is available. But let's go into training and fine tuning our own personal model. The last thing I'm going to show you on the left side, you have personal feed. But what we want to do is we want to click on training and data sets. Now, when I do that, I'm going to click on a new data set. I'm going to drag in what I'm going to do is I'm going to create a data set. And then I'm going to use this data set as the basically the images that go into my model. So I'm just going to call it new. I'm just going to make it real quick new and new, new, just because I just want to go quickly. Create data set. When you do that, you can now drag and drop files to upload or include here. Now you need to drop at least eight images in for it to go ahead and start creating a model. So I'm going to go to my finder and then I've already got a pre-selected 10 here. So I'm going to take these 10 images, drag and drop them. And these are all basically images done in the style of Caravaggio. He was an old school Italian artist and I really like his style, right? Really dark, lots of use of nice shadows. OK, so I've dragged and dropped them in there. Now on the right side, now that I've got my images, I'm going to go train model. And when I do that, I'm just going to go ahead and type in some info. So I'm just going to call it new. And I'm going to call it new. I'm just going really quickly through this because I don't want to waste your time and new instance prompt. OK, there we go. Now I'm going to go ahead and start training. When it does that, it says now the training is in progress. It's using those images. And then once it's done, it's going to email me. I have done this already a couple of times. So I know it takes about one minute for it to go ahead. Once it's done, you can then go into fine tuned models. And then here, these were the platform models, the ones that we looked at earlier. These are some community models that everybody, like a lot of people have made themselves that you can just go ahead and use. There's basically, you know, if you like them, use them. Isometric one is really good, by the way. But the one that I'm interested in is your models. These are the ones that I've made personally. So this is the one I'm making right now. There's one on underway ABC, I believe it's called. But here's an example of one that I just made. It's Caravaggio and it uses that style. So I'm going to click on view and look at that. This is my personal model and I've created 20 images with it. It uses stable diffusion 1.5. I'm just going to click on generate with this model. And then you're going to see here that it's using the fine tuned model. Caravaggio and now I'm going to create, if I created, if I just clicked on Bulldogs playing poker, it's using my model. And then in a few seconds, it's going to have some cool looking Bulldogs, but in the Caravaggio style that I created. Let's see what we get here. I'm going to create four of them. Here we go. So this is the Caravaggio style. This is the style that I created. They're a little low resi, but this one's pretty good actually, but there's no Bulldogs. Okay, this one here, this one's pretty good. That Bulldogs been to the gym. And then let's take a look at this one. It's all Bulldogs. Okay, that's a little bit dark, but like I said, remember when we upscale, we get a better shot of it. So we'll probably get better facial features once the upscaling is done. So I'm going to go ahead and click on that one. I'm going to take a quick look at this one. I like everything in this image except for the Bulldog. The Bulldogs done kind of a little bit weird. And then this one here, great image, but there's no dogs in it. So that's an issue. So let's just wait for this to finish. And then once that's done, I'm going to go ahead and show it to you. And we'll call it a day. All right, welcome back. So it is done. Let's click on it. Okay, the facial features are much better. Let's go with the original. What the heck is going on here? It's kind of generic. And then the upscaled version, so much better. It's in the Carvaggio style, a style I made. That's how you use this free tool, Leonardo AI, to make free images. Thanks for watching. A ton more stuff coming up. Stay tuned.