Stable Diffusion… The name has a nice ring to it, right? When it was first released, everyone was pretty overjoyed with it and rightly so. After all, this was the very first open sourced A.I. image generator. The creators, Stability AI, very generously handed this awesome tool of future to the entire world of creative minds and exceptional brains to play with and create more awesomeness.
But initially it was raw data and coding excellence, unattainable by regular masses without any coding knowledge. That’s where our saviors came in various formats. They created several platforms to access this brilliant tools. Some are free like playground, bluewillow etc, some are paid like runpod, rundiffusion and some work on a hybrid model like Leonardo AI.
But today I am gonna talk about Automatic 1111, the open sourced webui that gives the power in the hands of masses to install and run stable diffusion locally in their windows pc and explore the full range of its capabilities.
Here are the top 5 reasons why you should definitely learn to use Stable Diffusion :
1. Large library of custom models
There is no need to be tied to the models that are released by Stability AI. In a website like CivitAI.com you can find hundreds of custom models catered to different tastes and needs. Many people are developing their own models and kindly making them open sourced so that everyone around the world use that. It opens up an amazing possibility to get different models for your exact requirements and a large all you can eat buffet.
2. Train your custom models
In SD you can train your own custom models. If none of the custom models availbe online fancy you, you can always train your own custom model using the Dreambooth extension and even train them on your face with as low as 8 images. Although I suggest to use at least 20 images for the best results possible.
3. Controlnet
Controlnet is the single biggest thing that has happened to SD since its first release. With this extension you can attain any pose for a person, convert any architectural drawings into full fledged building drawings and what not. It is not really possible to explain about it in such a small post. I suggest you go look about it online, there are some really awesome contents to help you out there.
4. Inpainting
Inpainting gives an user to change any particular part of an image without changing it entirely. It can be used to enhance your images, add objects to it, change the background and much more. Now with adobe generative ai you can do the same but you need an adobe subscription for that whereas stable diffusion is completely free. And last but not the least you have a large library of models availbale out there.
5. Open Sourced Extensions
We have only just explored a small realm of possibilities with stable diffusion as of yet. Every day the worlds genius minds are working on developing new extensions that makes this tool better and amazing. Very generously they are making those tools open sourced and giving everyone the opportunity to explore all the possibilities at an unimaginable rate. Dreambooth, Controlnet, Lora, Depth Map, the list goes on and on. And to be honest each one of them are pretty amazing. If you are working with Generative AI tools, it just feels criminal to not explore what Stable Diffusion can do and make yourself limited.
If you want to get started with stable diffusion and don’t know where to start, here is an installation guide that beautifully explains how you can install the Automatic 1111 in your PC without any prior knowledge of coding and jump into the world of unimaginable. Click here to watch the video
