AI processing moves onto users' devices
Opera has introduced WebAI, a web standard designed to run AI models directly in a browser rather than relying on cloud services. The system is built to work with a range of open-weight models, including Google's Gemma. On Mac computers, it can also use Apple's on-device foundation model.
The feature is available as an opt-in early preview in Opera One. Opera says its Opera GX and Opera Neon browsers are expected to receive the capability in the future.
WebAI can support tasks such as generating text, classifying content and identifying relationships between information. Because processing takes place on the user's device, these tasks can be performed without sending the relevant data to a company's cloud servers. Open-weight models can also give developers flexibility to adapt AI features without operating their own AI infrastructure.
Local processing relies on a computer's available hardware, so performance may vary between devices. Opera's approach points toward a browser experience where more AI features can run on-device, giving users and developers greater choice over how those tools are used.
