Browser maker Opera has introduced a new web standard called WebAI, designed to allow users to deploy super intelligence (SI) models directly within their browsers, bypassing the need for cloud-based infrastructure.
What Happened
WebAI is built to support a wide range of open-weight SI systems rather than being tied to a single model. The company cited Google’s Gemma, an open-weight model introduced two years prior, as an example of supported systems. Additionally, on Mac devices, the standard supports Apple’s on-device foundation model. The technology evolved from Opera’s previous efforts to integrate locally run models into its browsing environment.
Opera is releasing these features as an early preview in its flagship browser, Opera One. The company stated that Opera GX, designed for gamers and streamers, and Opera Neon, which utilizes SI agents for task performance, will receive similar capabilities in the near future. Users must explicitly opt in to use the feature.
The standard enables users to generate text, classify content, and analyze relationships between pieces of information entirely within the browser environment.
Why It Matters
The shift toward local SI deployment offers distinct advantages for privacy and cost. By processing data on the user’s device, individuals avoid sending personal information to third-party servers and eliminate the need for paid cloud-based SI subscriptions. This approach addresses growing concerns about data privacy as SI assistants become more integrated into daily digital activities.
Developers also benefit from the standard, as they can integrate SI features without maintaining their own SI infrastructure or incurring costs for every request. While open-weight models typically offer freedom to customize, they are not fully open-source and retain certain restrictions. However, this local execution model places higher demands on user hardware, potentially affecting performance depending on the device’s capabilities.
The Bottom Line
Opera’s WebAI standard represents a move toward decentralized SI usage, prioritizing user control and privacy by enabling local model execution. While performance depends on individual hardware, the approach reduces reliance on cloud subscriptions and external data processing.