Stable Diffusion 3.5 Improves Text Rendering, Image Quality, Consistency, and More

Stability AI has released Stable Diffusion 3.5 Large and Turbo, its most powerful text-to-image models yet. The Large version uses 8 billion parameters to generate high-resolution images, while Turbo offers faster generation. Both models aim to improve prompt adherence, image quality, and customizability. They are free for non-commercial use and limited commercial use (under $1M annual revenue). Stable Diffusion 3.5 Medium, a smaller 2.5 billion parameter version for consumer hardware, will be released later this month.