Gradio 3.0 is Out!

Machine Learning Demos

Machine learning demos are an increasingly vital part of releasing a model. Demos allow anyone — not just ML engineers — to try out a model in the browser, give feedback on predictions, and build trust in the model if it performs well.

More than 600,000 ML demos have been built with the Gradio library since its first version in 2019, and today, we are thrilled to announce Gradio 3.0: a ground-up redesign of the Gradio library 🥳

What's New in Gradio 3.0?

🔥 A complete redesign of the frontend, based on the feedback we're hearing from Gradio users:

• We've switched to modern technologies (like Svelte) to build the Gradio frontend. We're seeing much smaller payloads and much faster page loads as a result!

• We've also embranced a much cleaner design that will allow Gradio demos to fit in visually in more settings (such as being embedded in blog posts).

• We've revamped our existing components, like Dataframe to be more user-friendly (try dragging-and-dropping a CSV file into a Dataframe) as well as added new components, such as the Gallery, to allow you to build the right UI for your model.

• We've added a TabbedInterface class which allows you to group together related demos as multiple tabs in one web app

Check out all the components you can use on our (redesigned) docs 🤗!

🔥 We've created a new low-level language called Gradio Blocks that lets you build complex custom web apps, right in Python:

Why did we create Blocks? Gradio demos are very easy to build, but what if you want more control over the layout of your demo, or more flexibility on how the data flows? For example, you might want to:

• Change the layout of your demo instead of just having all of the inputs on the left and outputs on the right
• Have multi-step interfaces, in which the output of one model becomes the input to the next model, or have more flexible data flows in general
• Change a component's properties (for example, the choices in a Dropdown) or its visibilty based on user input