In today’s post, I will discuss an intriguing AI tool, furry diffusion.
But first, we need to explain Stable diffusion, as furry diffusion is just a model of this AI image-generating tool.
Stable Diffusion is a machine learning model that may be used to generate pictures from text, fill in details on low-resolution photos, and edit images depending on the text. You may already be familiar with this concept, as it has the potential to come to your attention.
This technology is becoming more popular not just among traditional movie and television production companies but also among independent content producers working on different social platforms.
What is the Mechanism Behind Stable Diffusion?
This model, similar to Google’s image, conditions the model on prompts by using a frozen encoder text (CLIP ViT-L/14).
To begin, we will check to see whether your personal computer is capable of running Stable Diffusion. It needs a dedicated GPU memory of 4 GB, and you must also ensure that you have 10 GB of free space on your hard drive. In order to make use of it, we need to acquire the following two prerequisites:
Git is often used for the maintenance of source control, but in this instance, we will make use of it in order to download Stable Diffusion and also to stay current. (Install GIT; we will need it in the future.)
Stable Diffusion is developed in Python, which is a programming language. Python is also the name of the game.
How Does One go About Setting Up Furry Diffusion?
You can find a detailed tutorial with specific steps for local installation.
You may get a direct connection to AUTOMATIC-1111’s WebUI on github.
This distribution simply contains the user interface tool. Download one of the models from the “Model Downloads” section, rename it to “model.ckpt,” and then store it in the /models/Stable-diffusion folder on your computer. This will allow you to use it with a custom model.
Using Windows with an AMD graphics processing unit
- Epoch 18 General-use model that was trained on e621 is now available for download.
DURING THE TRAINING, the word “explicit” was spelt as “explict,” therefore please note this.
Zack3D – Kinky Furry CV1
- An application that specializes on goo and latex but is also capable of producing high-quality general furry art and is friendly to NSFW.
They have a new channels section, one for AI generations! In there they have the paid midjourney bot, and the free-for-you custom stable diffusion bot. The stable diffusion bot is a project of mine, and is able to generate really good furry art of all kinds! Zack the discord creator expressively wrote “A couple images from that bot have even ‘escaped’ and are floating around rubber groups, so you may have seen an image from it and never known”. They incourage you to try the bot for free but there is a reminder that says “Please do try it out, but just know it is not cheap to run (over 200usd a month!)”.
In the discord down below you can find midjourney-bot, stable-diffusion-bot, stable-diffusion-bot-2, dalle2-share. Making the channel very entertaining for Art lovers.
Diffusion of Ponies
- By the process of fine-tuning, the latent text-to-image diffusion model known as pony-diffusion has been conditioned on high-quality pony SFW-ish pictures.
Information on the model and links to downloads, whereas the person’s that created Diffusion of Ponies – Twitter.
Examples of Furry Stable Diffusion
They identify with the e621 definition (an animal character that is depicted in its natural form. The art form mostly is generated on ferals that usually move on four legs, but there are categories that incldue naturally bipedal animals (images) kangaroos, birds, and various dinosaurs.
In their discord channel you can also find a 18+ section, in which people generate explicit images. They have around 15.000 members, and most of them are very active.
In my opinion this is a great community as far as you enjoy the diversity of images generated, and even for those that want to find themselves comfortable within a community that enjoys explicit images to a certain degree.
In the following you can see some non-explicit images generated by Furry Diffusion AI: