Sorry to be the one to break it to you, but if you're currently at a point where you can't even Google how to do it, it's going to feel impossibly hard to learn without someone holding your hand through the process.
First of all, animations are just sequences of images, so you should first learn how to generate those.
Look at stable diffusion implementations like automatic1111, which is the OG, but there's a "better" fork of this called SDNext with faster updates and probably more features. (You said there's nothing in Google, but that's a lie—there's a lot. You just don't know how to search.)
I'd personally advise you to use ComfyUI. You'll understand the process much more by learning it and ultimately have much more control and flexibility. (I'm pretty sure it's also supported or at least recommended by Stability AI).
Keep in mind that to run it locally, you need a decent GPU, but to use it and not lose your mind, you need a good GPU.
For example, right now I have something which I consider in between these two categories (RTX 3070), and I left generating images until I get a 4080 Super or 4090 in a few months. (It just took too much time).
On RTX 3070 with 8GB of VRAM, you already have big VRAM issues with XL models (which are currently the newest). It also takes way too much time to generate those animations with my RTX 3070, so if you have anything worse, you better be really patient.
I'm not even going to go into issues related to generating animations by themselves, like probably the biggest issue, which is to keep the frames consistent.
By the way, I don't want to discourage you from doing this, but I want to temper your expectations.
