SDXL Turbo is a large Vincent graph model released by Stability AI. SDXL Turbo uses groundbreaking Adversarial Diffusion Distillation (ADD) technology to set a new standard in text-to-image generation. SDXL Turbo class
SDXL Turbo is a large Vincent graph model released by Stability AI. SDXL Turbo uses groundbreaking Adversarial Diffusion Distillation (ADD) technology to set a new standard in text-to-image generation.
The number of steps required for SDXL Turbo to generate images similar to LCM has been changed from the original 50 steps to 1 step. According to their CEO, SDXL Turbo can currently achieve 14 frames per second image generation on the 4090.
Currently on their official website, you can try SDXL Turbo directly for free (the generated image has an SDXL Turbo watermark in the lower right corner)

Briefly describe the image content you want to generate, and then select a style model. No registration or login required, free users can generate two pictures at the same time. Take a look at the following demonstration. Is it very realistic?

If you don’t know how to write prompt words, you can find images generated by other users in AI Image Prompts. On the details page, you can see the model used for the image and the prompt words. You can imitate or directly copy the prompt word.

In addition to free trials of SDXL Turbo models, the website also supports the use of Stable Diffusion 3 models, but they can only be generated twice a day.
