Stable Diffusion XL Turbo generates AI images as fast as typing

Stability AI has launched Stable Diffusion XL Turbo, an AI image-synthesis model that can rapidly generate imagery based on a written prompt. The 'real-time' image generation can transform images from a source quickly.

The XL Turbo software can produce image outputs in a single step, a significant reduction from the 20–50 steps required by its predecessor. The model uses Adversarial Diffusion Distillation (ADD) which learns from existing image-synthesis models, and adversarial loss, enhancing the model's ability to differentiate between real and generated images, and improving the realism of the output.

The SDXL Turbo images aren't as detailed as SDXL images which won't be replaced in its results but takes longer to generate images than Turbo. Blind tests conducted with the programme revealed that SDXL Turbo could deliver superior results compared to the LCM-XL model in just one step instead of four.

The most fascinating feature about the software is, that it generates images while you type, changing the image as you type in more details.

SDXL Turbo is available under a non-commercial research license, limiting its use to personal, non-commercial purposes. Try out the new generative model on Stability AI's image editing platform, Clipdrop. The software supports all browsers and is free for users in Beta phase.