Expanding access to Opal, our no-code AI mini-app builder (blog.google)

Google’s no-code AI mini-app builder Opal — launched two months ago in the U.S. as a Google Labs experiment — is expanding to 15 more countries, including Canada, India, Japan, South Korea, Brazil, Singapore and several Latin American markets. The move responds to unexpectedly sophisticated user creations and aims to broaden access to a tool that lets creators build AI-powered workflows with natural language rather than code. To support wider, more complex usage, Opal is rolling out two major upgrades. First, an advanced, no-code debugging experience: you can run workflows step-by-step in a visual editor or iterate on individual steps in a console panel, with real-time error reporting localized to the exact failing step. Second, core performance improvements — dramatically faster app creation (previously up to five seconds) plus parallel runs that let multi-step workflows execute simultaneously — which reduce latency and improve throughput for complex automations. For the AI/ML community these changes lower the barrier to iterating on and operationalizing models and pipelines, reduce debugging friction typical in MLOps, and help scale richer, production-ready AI mini-apps across new international user bases.