Opera Launches WebAI Standard to Enable On‑Device AI in Its Browser

What Is WebAI?

Opera has unveiled a new web standard called WebAI that lets users run AI models directly inside the browser, bypassing the cloud. The specification is model‑agnostic and already demonstrated with Google’s Gemma—an open‑weight model Google first introduced two years ago—and Apple’s on‑device foundation model on macOS. WebAI grew out of Opera’s earlier work to bring locally executed models into its browsers.

Availability and Opt‑In

The feature is offered as an early preview in the flagship Opera One browser. Opera plans to extend the same capability to its gaming‑focused Opera GX and AI‑driven Opera Neon browsers soon. Users must explicitly opt into the feature to use it.

Benefits of Open‑Weight Models

• Usually free to download and use.
• Model weights are publicly available, giving developers flexibility to customize behavior.
• Privacy is enhanced because data never leaves the user’s device.
• Developers can integrate AI without maintaining server infrastructure or paying per request.

Unlike fully open‑source software, open‑weight models still carry some usage restrictions.

Privacy Considerations

Privacy has become a major concern as AI assistants gain access to personal information. Running AI locally keeps processing on the user’s machine, giving users greater control over how their data is handled, though it does not eliminate all risks. Opera has long positioned digital privacy as a core selling point.

Capabilities

WebAI enables tasks such as text generation, content classification, and relational analysis entirely within the browser.

Performance Trade‑Offs

Executing AI models on a local device increases CPU and memory usage, and actual performance will vary according to the hardware. Opera notes that users should expect slower results on less powerful machines.