arXiv + Replicate – Replicate

Hello, arXiv visitor 👋

Welcome to Replicate. We're making it easy to share your machine learning model. Check out some recently featured models:

Edit images with human instructions

A latent text-to-image diffusion model capable of generating photo-realistic images given any text input

Bach chorale generation and harmonization

Replicate is currently in closed beta while we iron out the wrinkles. Tell us about your model and we'll add you to the waitlist.