Stability AI releases Stable Diffusion 3.5 with enhanced speed, quality, and efficiency

Stability AI has unveiled Stable Diffusion 3.5, featuring multiple model variants including Stable Diffusion 3.5 Large and Stable Diffusion 3.5 Large Turbo. The Stable Diffusion 3.5 Medium model is set for release on October 29th.

Stable Diffusion 3.5 Large, with 8 billion parameters, is the most powerful model in the series, offering superior quality and prompt adherence at 1 megapixel resolution, making it suitable for professional use. The Stable Diffusion 3.5 Large Turbo, a distilled version, provides high-quality images in just 4 steps, significantly faster than its counterpart. The upcoming Stable Diffusion 3.5 Medium model, at 2.5 billion parameters, features an improved MMDiT-X architecture and is designed for consumer hardware, capable of generating images between 0.25 and 2 megapixels.

All models are highly customizable, run on consumer hardware, and are free for commercial and non-commercial use under the Stability AI Community License. They excel in customizability, performance efficiency, and output diversity.

Stable Diffusion 3.5 Large and Large Turbo are available for download on Hugging Face, with model weights accessible for self-hosting. The models can also be accessed via Stability AI API, Replicate, ComfyUI, and DeepInfra.