Cog example models

This repo contains example machine learning models you can use to try out Cog.

Once you've got a working model and want to publish it so others can see it in action, check out replicate.com/docs.

Examples in this repo

• hello-world - Takes a string as input and returns a string as output. The simplest possible "model".
• resnet - Classifies images using ResNet. This is a good example of a deep learning model that's small enough to run without a GPU if you need to do that for demos.
• z-image-turbo - Z-Image-Turbo image generation model. This is an example of a modern generative model. It requires a GPU, but it's small enough that it's fast to run, so makes for a good demo.
• blur - Applies box blur to an input image
• canary - Takes a string as input and returns a streaming string output
• notebook - Using a Jupyter Notebook with Cog
• hello-train - Demonstrates Cog's training API for fine-tuning