Available now

Illusion Diffusion: text to image

Hide a high-contrast shape or word in a landscape or illustration. Illusion Diffusion uses your reference for the hidden outline and your prompt for the scene around it.

Model ID

Input

Fields marked * are required. Review the price before generating.

Uploading a file needs an account. Create account

JPEG, PNG, WebP · up to 25 MB

Be as descriptive as possible for best results.

Example prompt. Edit it or write your own.

You can choose between some presets or custom height and width that **must be multiples of 8**.

Additional settings

It will be the same value of the one passed in the input or the randomly generated that was used in case none was passed.

Scheduler / sampler to use for the image denoising process.

Price$0.60 / run

You sign up first. Your settings are kept.

About this model

Overview

Upload a high-contrast image, such as a spiral, a logo or black text on white, and describe a scene. Illusion Diffusion builds the scene so its light and shapes follow your image. A ControlNet scale sets how strongly the pattern shows, and the start and end values set when it guides the render.

What you can build

  • Landscapes with a hidden word or logo
  • Spiral and checkerboard illusions in a city or forest
  • Posters that reveal a shape when seen from afar

Prompting tips

  • Use simple black and white input images with bold shapes; fine detail gets lost.
  • Raise the ControlNet scale when the pattern disappears, lower it when the scene looks flat or forced.
  • Pick scenes with lots of small elements, such as villages, crowds or foliage, so the pattern has material to form from.

What to check

Check whether both the pattern and the surrounding scene remain readable at the intended display size.

Inputs

  • Image (required): Input image url.
  • Prompt (required): The prompt to use for generating the image. Be as descriptive as possible for best results.
  • Negative Prompt (optional): The negative prompt to use. Use it to address details that you don't want in the image. This could be colors, objects, scenery and even the small details (e.g. moustache, blurry, low resolution).
  • Guidance Scale (optional): The CFG (Classifier Free Guidance) scale is a measure of how close you want the model to stick to your prompt when looking for a related image to show you.
  • Controlnet Conditioning Scale (optional): The scale of the ControlNet.
  • Control Guidance Start (optional) From 0 to 1.
  • Control Guidance End (optional) From 0 to 1.
  • Seed (optional): Seed of the generated Image. It will be the same value of the one passed in the input or the randomly generated that was used in case none was passed.
  • Scheduler (optional): Scheduler / sampler to use for the image denoising process. Options: DPM++ Karras SDE, Euler.
  • Number of inference steps (optional): Increasing the amount of steps tells Stable Diffusion that it should take more steps to generate your final result which can increase the amount of detail in your image. From 0 to 80.
  • Image Size (optional): The size of the generated image. You can choose between some presets or custom height and width that **must be multiples of 8**. Options: Square HD, Square, Portrait 4:3, Portrait 16:9, Landscape 4:3, Landscape 16:9.

Output and limits

  • Image sizes: Square HD (1024×1024), Square (512×512), Portrait 4:3 (768×1024), Portrait 16:9 (576×1024), Landscape 4:3 (1024×768), Landscape 16:9 (1024×576).
  • Small or dense text in the image can come out misspelled.
  • Hands, crowds and complex poses can show artifacts; generate a few variations and pick the best.
  • Some prompts or outputs can be refused by safety checks.

Price guide

  • up to: $0.60 / run