Stable Diffusion 1.5 (accelerated)
A deep learning model capable of generating photo-realistic images based on text input. You can run the world’s fastest and cheapest Stable Diffusion endpoint today in OctoAI.
Advice from ML Experts
Different samplers can give slightly different perceptual qualities to your images and can vary in the number of steps to create a good image. The "DPM" family of schedulers can get good images in 20 - 25 steps so can be used for quick iteration and prompt tuning. By saving the seed used to generate your initial images, you can always rerun the same job with a higher number of steps to get a more refined image.
Input
Output
Generated in: XX ms
Run Time & Costs
OctoAI brings you the fastest Stable Diffusion available today. Internal benchmarking, run against creation of 512x512 images with 30 steps shows image generation latency under 1.5 seconds when running on the A10G tier, and under 0.7 seconds on the A100 tier. Detailed benchmarking is in progress, and additional performance data will be updated here once available.
Model Information
Developed by: Robin Rombach, Patrick Esser
Model type: Diffusion-based text-to-image generation model
Language(s): English
License: The CreativeML OpenRAIL M license is an Open RAIL M license, adapted from the work that BigScience and the RAIL Initiative are jointly carrying in the area of responsible AI licensing. See also the article about the BLOOM Open RAIL license on which our license is based.
Model Description: This is a model that can be used to generate and modify images based on text prompts. It is a Latent Diffusion Model that uses a fixed, pretrained text encoder (CLIP ViT-L/14) as suggested in the Imagen paper.
Resources for more information: GitHub Repository and Paper.
Cite as:
@InProceedings {Rombach_2022_CVPR,
author = {Rombach, Robin and Blattmann, Andreas and Lorenz, Dominik and Esser, Patrick and Ommer, Bj\"orn},
title = {High-Resolution Image Synthesis With Latent Diffusion Models},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2022},
pages = {10684-10695}
}