Google has introduced Opal , a new application for creating and testing programs that operates in the browser and utilizes AI capabilities of Gemini. Opal is aimed at developers, students, and enthusiasts who need a simple tool for writing code with AI support. Users can run code, receive explanations, and instantly get debugging help and suggestions directly in the workspace.
🚨 NEW LABS EXPERIMENT 🚨
— Google Labs (@GoogleLabs) July 24, 2025
Introducing Opal, our new way to help you build and share AI mini-apps by linking together prompts, models, and tools— all while using simple, natural language (without a single line of code 🤯)
Now available in US-only public beta! Learn more ⬇️… pic.twitter.com/9Csk1EruJu
Opal allows creating applications using natural language and interactive visual prompts. The app includes a gallery of ready-made templates that can be used for various purposes, such as creating virtual games or video ads. For those who want to work from scratch, there is a window for entering their own instructions, which the system converts into a step-by-step visual scheme of the program’s operation.
Opal supports several programming languages and combines the capabilities of Gemini, Veo, and Imagen models in one interface. Gemini assists with text and explanations, Veo creates videos with audio, and Imagen generates images for applications. All stages are displayed as separate blocks that can be viewed or edited.
In the upper right corner of Opal, there is a “Share app” feature that allows sharing created applications via links with other Google users. The app is currently in open beta for users in the US, and Google plans to expand access after evaluating feedback and activity from the first wave of users.
