Opera One Adds Local AI Model Support via WebAI
The new standard allows browser extensions to execute AI models on local hardware, bypassing cloud-based subscription requirements.
Updated on Oct. 27, 2026 in Artificial Intelligence
Live Poll
Would you prefer your browser to run artificial intelligence models directly on your own computer?
Opera One has introduced the WebAI standard, enabling browser extensions to run artificial intelligence models directly on user hardware. This feature is currently live, allowing users to opt into local execution of models like Google’s Gemma and Apple’s on-device foundation model without recurring subscriptions.
Why it matters
Running models locally shifts the compute burden from cloud-based AI infrastructure to the user device, bypassing data latency and subscription fees. This development signals a broader transition toward privacy-centric, on-device AI as browsers increasingly serve as local execution runtimes.
The WebAI standard serves as the interface between the Opera One browser and local hardware, supporting models including Google's Gemma and Apple's foundation model. This approach moves model inference away from external servers, reducing infrastructure costs for the user.
The players
Opera
A software company known for its web browser suite, which includes Opera One, Opera GX, and experimental projects like Opera Neon.
A major technology firm that developed the Gemma open-weights AI model architecture.
Apple
A hardware and software company that designs the on-device foundation models utilized by the browser.
The details
The WebAI standard functions as an intermediary layer that allows browser extensions to tap into the device's local compute resources. By keeping processing on the client side, the browser removes the dependency on cloud infrastructure, which typically handles the heavy lifting of model inference. Additionally, the new HTML-in-Canvas feature provides developers with the ability to render functional web pages into 2D and 3D scenes using standard HTML and CSS primitives.
Timeline
- 2026-10-27
Opera released the WebAI standard and HTML-in-Canvas functionality.
The Tech Race
Opera's integration of models like Google's Gemma signals a move to make local LLM execution a standard browser capability. This effort competes with efforts by other major browsers that still predominantly rely on cloud-bound model queries for AI features.
Users can access local AI capabilities immediately within Opera One without needing a paid subscription or external cloud account. Future updates will bring this same on-device functionality to the Opera GX and Opera Neon browser versions.
The takeaway
The transition to on-device AI removes the cost and connectivity barriers typically associated with high-performance model access. Users should monitor Opera’s future announcements regarding the expansion of this local model support to the Opera GX and Opera Neon product lines.
Further reading
For more on how browsers are evolving to host neural networks, visit the Artificial Intelligence section.
Live Poll
Would you prefer your browser to run artificial intelligence models directly on your own computer?
