Prototype with Gemini models, structured output, code, and multimodal inputs in a browser developer workbench.
Coding & Dev

Gemma 4 is Google's open multimodal model family for building compact applications that understand text, images, and audio.
Best for: Building on-device AI prototypes, Testing multimodal understanding, Fine-tuning an open model for a niche task
Prototype with Gemini models, structured output, code, and multimodal inputs in a browser developer workbench.
Prompt an in-browser coding agent to build, run, and host a full-stack web application.
Work with an AI coding agent that can inspect files, edit code, run commands, and test changes with your approval.