1 min read
Add as a preferred source on Google

Opera Introduces WebAI Standard for Running AI Models in Its Browser

The opt-in feature supports multiple models, including Google’s Gemma and Apple’s on-device foundation model on Macs.

A laptop displays simple browser cards beside a disconnected cable / TokenPost.ai
A laptop displays simple browser cards beside a disconnected cable / TokenPost.ai

Opera introduced WebAI Tuesday, a browser standard for running AI models on a user’s device, giving people access to AI features without relying on cloud processing.

The system supports multiple models. Opera named Google’s Gemma as one option and said Mac users can also use Apple’s on-device foundation model. WebAI is designed to accommodate a range of open-weight models, whose publicly available weights give developers more freedom to customize them, though the models may still carry restrictions.

The initial preview is available in Opera One, the company’s flagship browser. Opera plans to bring the feature to Opera GX, its browser for gamers and streamers, and Opera Neon, its browser that uses AI agents, soon.

WebAI can generate text, classify content and identify relationships between pieces of information. Developers can add these capabilities without maintaining AI infrastructure or paying for each request. Users must opt in, and information handled by the local AI remains on their device. The features do not require a cloud AI subscription.

Local models use a computer’s processing power, so performance depends on the device’s hardware.

Simon Yoon

Reporter

Simon Yoon reports on blockchain technology for TokenPost. Send corrections or tips to info@tokenpost.com.

Loading…