lucataco / demofusion

DemoFusion: Democratising High-Resolution Image Generation With No 💰

  • Public
  • 9.2K runs
  • A100 (80GB)
  • GitHub
  • Paper
Run with an API
Playground API Examples README Versions

Input

string
Shift + Return to add a new line

Input prompt

Default: "An astronaut riding a rainbow unicorn"

string
Shift + Return to add a new line

Input Negative Prompt

Default: "blurry, ugly, duplicate, poorly drawn, deformed, mosaic"

integer

Width of output image

Default: 3072

integer

Height of output image

Default: 3072

integer
(minimum: 1, maximum: 500)

Number of denoising steps

Default: 50

number
(minimum: 1, maximum: 50)

Scale for classifier-free guidance

Default: 7.5

integer

The batch size for multiple denoising paths

Default: 16

integer

The stride of moving local patches

Default: 64

number

Control the strength of skip-residual

Default: 3

number

Control the strength of dilated sampling

Default: 1

number

Control the strength of the Gaussian filter

Default: 1

number

The standard value of the Gaussian filter

Default: 1

boolean

Use multiple decoders

Default: true

integer

Random seed. Leave blank to randomize the seed

Add a payment method to run this model.

Each run costs approximately $0.0073. Alternatively, try out our featured models for free.

Sign in with GitHub

By signing in, you agree to our
terms of service and privacy policy

Output

outputoutputoutput
Generated in
Tweak itReport

This example was created by a different version, lucataco/demofusion:6195e015.

Examples

View more

Run time and cost

This model costs approximately $0.0073 to run on Replicate, or 136 runs per $1, but this varies depending on your inputs. It is also open source and you can run it on your own computer with Docker.

This model runs on Nvidia A100 (80GB) GPU hardware. Predictions typically complete within 6 seconds. The predict time for this model varies significantly based on the inputs.

Readme

This is an implementation of DemoFusion. DemoFusion provides highres txt2img capabilities based on SDXL. See the demo example that showcases a txt to img run that provides an img in 1024x1024, 2048x2048, and 3076x3076 resolution in under 6min

Abstract

High-resolution image generation with Generative Artificial Intelligence (GenAI) has immense potential but, due to the enormous capital investment required for training, it is increasingly centralised to a few large corporations, and hidden behind paywalls. This paper aims to democratise high-resolution GenAI by advancing the frontier of high-resolution generation while remaining accessible to a broad audience. We demonstrate that existing Latent Diffusion Models (LDMs) possess untapped potential for higher-resolution image generation. Our novel DemoFusion framework seamlessly extends open-source GenAI models, employing Progressive Upscaling, Skip Residual, and Dilated Sampling mechanisms to achieve higher-resolution image generation. The progressive nature of DemoFusion requires more passes, but the intermediate results can serve as “previews”, facilitating rapid prompt iteration.

@article{du2023demofusion,
    title={DemoFusion: Democratising High-Resolution Image Generation With No $$$},
    author={Ruoyi Du and Dongliang Chang and Timothy M. Hospedales and Yi-Zhe Song and Zhanyu Ma},
    journal={arXiv},
    year={2023}
}