Replicate AI Face-to-Sticker

What is Replicate AI Face-to-Sticker?

Face-to-Sticker is a model hosted on Replicate that turns faces into stickers. You provide a facial image and adjust a few settings — width, height, a descriptive prompt, and other options — to shape the final sticker. It’s a simple way to turn photos into personalized stickers for creative projects.

Who created it?

The model was created by fofr. The special Loras that power it were developed by artificialguybr, who accepts support via Patreon and Ko-fi and can be followed on Twitter. The model runs on Nvidia A40 (Large) GPU hardware.

How to use it

There are several ways to run fofr/face-to-sticker:

1. Replicate’s API: Run the model directly through the API. Inputs include the image URL, number of steps, width and height, a style prompt, plus settings like upscale amount and prompt strength. See the model’s schema for the full list of inputs and outputs.
2. Docker: Pull and run fofr/face-to-sticker locally using Docker, which downloads the full model to your machine. A specific command is provided to get it running.
3. Cog: Alternatively, pull and run the model with Cog. Inputs are set up the same way — image URL, steps, width, height, prompts, and other parameters.
4. Examples: Visit the model page to see more examples of how people use the tool for different creative projects.
5. API token: To authenticate and get predictions via Replicate’s API, you’ll need your API token, found in your account settings.

Whether you run it through the API, Docker, or Cog, the workflow is the same: supply a face image, tune the parameters, and generate a sticker.