Opera has introduced WebAI, a new web standard that allows users to run AI models directly in their browsers without relying on cloud-based services.

 

WebAI supports a range of AI models, including Google’s Gemma and Apple’s on-device foundation model on Macs. Opera says the technology is not tied to a specific model and is designed to work with various open-weight AI systems.

 

The feature is currently available as an early preview in Opera One. Opera GX and Opera Neon are expected to receive support soon. Users must opt in to enable it.

 

Open-weight AI models offer several advantages, including free access in many cases and greater flexibility for developers to customize them. They can also improve privacy by allowing AI processing to happen directly on a user's device rather than sending personal data to external servers.

 

Opera says WebAI enables developers to integrate features such as text generation, content classification, and information analysis without maintaining AI infrastructure or paying for every cloud-based request.

 

The move comes as concerns about data privacy grow alongside the adoption of AI assistants. Running models locally gives users greater control over their information, although it does not eliminate every privacy risk.

 

However, local AI processing has limitations. Running models directly on a device requires sufficient computing power and memory, and performance will vary depending on the user's hardware.