Ultimate Stable Diffusion AI art Course (beginner to pro)

Congratulations on taking the first step in the world of AI art! This is your first lecture, and to ensure quick results, we're going to show you how to try out various AI image generation models for free. In the upcoming lectures, you will learn two ways to deploy Stable Diffusion in the cloud, and then finally, we'll get to local deployment for free, and how to troubleshoot any issues you may encounter along the way.

If you ever find that any of the links stop working, just send me a message, and I'll try to update it as soon as I can.

In this lecture, we're going to learn about a professional tool, which is also the main tool for this whole course. We'll start by showing you how to run it online, and then in section 4, we'll teach you how to install it locally for free.

If you encounter problems while using Colab, then the runpod introduced in this lecture will be a great alternative. I will show you how to run Stable Diffusion on runpod and demonstrate how to download other third-party models.

In this lecture, we're going to talk about how Diffusion models actually work. After this, you'll have a pretty good overview of the AI image generation process, which will help you learn faster and understand what all the different parts of the web UI do in later lectures. this is also setting the stage for the ComfyUI section we're going to add in the future.

If you understood the last lecture, then this image-to-image part will be really easy to grasp. These two lectures are very important, so don't skip them. If you're still confused about anything, feel free to send me a message : )

Before moving on, watch this video to get an overview of what's coming up. This will help you maximize your learning and get the most out of the course.

The prompt is one of the key elements at the heart of AI art. If you don't know how to write a proper prompt and simply resort to copying and pasting prompts from others, you might encounter several challenges on your journey to becoming an AI artist. As you may have heard, "Good Artists Copy; Great Artists Steal." AI art creation shares many similarities with traditional art, such as drawing inspiration, borrowing ideas, and most importantly-- practice.

While using the "git pull" command to keep the web UI up to date can be beneficial in terms of accessing new features and bug fixes, it's important to consider potential issues that might arise from new versions. Just like with the latest versions of Apple's iPhone software, there can be unforeseen bugs or compatibility issues.

We will talk about how to resolve some common issues encountered when deploying StableDiffusion locally.

If you find that the webui is generating images too slowly, or if you're encountering problems while deploying the webui, such as issues with git or python, you can use Forge, which we're introducing here, to solve these problems. This is because its one-click installation package already includes git and python. I hope this allows all of you to freely use Stable Diffusion to generate images!

After I finished recording this course, there was an update to the web UI of "Automatic1111." The interface now has some minor differences in appearance compared to what was covered in the course, but the core functionalities remain largely unchanged.

This lecture is mainly for those who have already used Stable Diffusion before. If you're a complete beginner, I recommend finishing the current section first before diving into this one. If you don't quite understand the VAE or Lightning model mentioned here, feel free to skip ahead and continue with the later content. I'll keep adding more lectures like this to help you learn more.

Want to significantly improve the speed and effectiveness of facial detailing? This section introduces the Adetailer plugin to help you do just that.

In this lecture we're going to use the flux model locally. Using the flux model is pretty similar to using regular Stable Diffusion models, it supports text-to-image, image-to-image, and inpainting. It's just that forge doesn't support controlnet for flux yet.

Introduce how to use online services and how to use newest SDXL base model and refiner model on Automatic1111, and compare the effect of refiner model.