 So you don't know how to use the best open source AI generator, not a problem. For the purpose of this series I will be using the official Stable Diffusion site. There are two main reasons for this, one I love what this AI stands for and I want to support them. And every time you buy credits here that money goes straight to the devs who then use that money to improve the product for everyone. So I'll leave a link to this site in the description for anyone who wants to use SD online. The second reason is I'm using this site because I want to keep this series accessible to the average dude. It's truly awesome that you can install this software locally but the truth is the overwhelming majority of people out there don't have a custom built PC. They don't know how to use GitHub or a command prompt and really don't have the time or resources to train an AI locally. So to keep this series easy for everyone we're just going to stick to the official website. It is paid but I'll leave a link to other websites you can use that are totally free. They're just a lot slower. Now when you get to the site it's pretty streamlined and it defaults to dark themes so you know it's legit. They have this super awesome width and height slider controller that allows you to change the dimensions of the image so if you're making a wallpaper you probably want it to be more horizontal and if you want pictures for mobile phone screens you probably kick that vertical up instead. CFG just means how literally it's going to try and follow your prompt. It defaults to 7 but if you set it to 0 it will basically give you completely unrelated images. If you set it to max it will give you the closest idea it can to your prompt word for word. But it also won't be as experimental or play with ideas very much. So 7 is usually a good amount to get what you want but also sometimes get stuff you didn't know you wanted. Steps is like how much extra time it spends diffusing your image. It's kind of hard to explain so I'll just show you. If you set it low it looks like this and the image is completed a lot faster. Now honestly it still looks pretty good but if you max it out all the way it takes way longer to generate but the images come out looking a lot more sophisticated. Number of images is just how many you get each time you generate. I have it set to 9 but you can set it to something like 1 or 4 if you want. And I'll be completely honest with you guys I have zero idea what Sampler does but I'll show you what all the settings look like and maybe you can leave a comment down below if you notice the changes. KLMS is the standard which I've been using this whole time. Here is KDPM2 Ancestral, KDPM2, K Hun, Cooler Ancestral, Cooler, PLMS and DDIM. So yeah I don't know how these affect the results but that's what they look like. If you like what you get you can download all the images with this button or download them all as a zip in a single file here. Now another thing that you can do which is kind of similar to Dolly is the image editor. Now there is a glitch where the tools do not appear if you're using Firefox so what I'm about to show you only works in Google Chrome, hopefully they'll fix that soon. But anyway if you go down here to the image editor you will be able to upload any image you want and once you're in the editor you can click here to scale your image or pan it around like this. If you click the brush you can erase parts of your image just like in Dolly. Brush size is controlled over here, the lower it is the smaller your brush the higher it is the bigger your brush. How sharp the brush is is controlled here so if it's down the brush is gonna be super sharp and if it's up it's gonna be super foggy. Also you can control Z if you make a mistake, strength controls opacity so all the way down makes the brush super weak and all the way up makes the erase really strong. Image opacity controls the transparency of the entire map but we will talk more about that later. Now there's another glitch here if you are brushing and your mouse goes outside the canvas it disables your brush and can be really annoying when you're trying to paint the edges because if you accidentally go out of bounds you kind of have to left click again. It's not a total deal breaker but hopefully they'll fix that soon. This is the opposite of the erase brush instead of erasing it restores your image back to its original state and this lets you see what the original image looked like and you can go back to the edited version if you just click on any of the other tools. Anyway just like a normal image you can change the width and height here which is a great easy way to make more space choose all your settings like normal, right your prompt and you're done. Now from what I've seen if you want to mutate an image that's when you use image opacity. So if we go back do this and then try the same thing you'll see how it starts to try and mutate the image slightly. The more transparent the more aggressive the mutation. So hope that helps and as always hope you have a fantastic day and I'll see you around.