Stable DiffusionUnclaimed

Stability AI's image models turn text prompts into images. The lineup includes Stable Diffusion 3.5 in three versions. Large is the most powerful model in the family, aimed at professional use at 1 megapixel resolution. Turbo is designed to run faster than Large and generates high-quality images in four steps. Medium strikes a balance between quality and customisation and is designed to run on consumer hardware.

The page also lists Stable Diffusion XL, a text-to-image model with 3.5 billion parameters for high-resolution, photorealistic outputs, and Stable Diffusion XL Turbo, a distilled version that can generate images in as few as one step.

According to the page, the models follow prompts closely and can produce a range of styles, including 3D, photography, painting and line art. They are also meant to produce images that represent a diversity of people and scenes around the world.

The image editing tools cover erasing objects, inpainting, outpainting, removing and replacing backgrounds, searching and recoloring or replacing items, upscaling images to larger sizes, and transforming images or sketches with control options.

There are four ways to access the models. You can deploy them on your own infrastructure with a self-hosted license, call them through the Stability AI API, use them in cloud partner ecosystems without extra infrastructure, or create images in Stable Assistant, a web-based creative platform.

The page addresses creators who want to make images without setting anything up, developers building custom applications, and enterprises that need to run the models in their own environment. The page does not list prices. Its buttons ask visitors to get a license or use the API, so pricing appears to be arranged directly with the company.