Opera opens a path for websites to run local AI through WebAI
Opera’s early-preview WebAI implementation lets websites and extensions use different AI models directly inside the user’s browser.
AT A GLANCE
Key points
- WebAI aims to let websites and extensions run AI models directly in the browser.
- Prompt, Decision, and Semantic Embeddings APIs are available in early preview.
- The feature is opt-in, with the initial preview limited to selected Windows and Apple-silicon systems.
What is changing?
Opera introduced WebAI, a proposed web standard, in early preview across Opera One, Opera GX, and Opera Neon on October 27. It allows websites and extensions to use AI models built into the browser without setting up a separate local server or obtaining an API key. Because processing remains on the device, content does not have to be sent to a cloud provider.
WebAI provides a Prompt API for text generation, a Decision API for classification and decision tasks, and a Semantic Embeddings API for finding relationships between content. Opera’s implementation supports open-weight models on its inference engine as well as compatible models already on the device. The feature is opt-in and downloads nothing until enabled. The preview supports 64-bit Intel/AMD Windows PCs and Apple-silicon Macs, but not Intel Macs, Linux, or Windows on Arm.
Explore the original sources
Source date: 27 October 2026.
Published by AI Radar: 27 Oct 2026, 16:10 (Istanbul, UTC+3).
FOLLOW THE TOPIC
Topics and tags
Share your discovery.
Send this story to a friend.
Reader comments
What do you think about this development? Share your experience or question. Comments are published after editorial approval.
Your name/nickname and comment become public if approved. Read: Privacy notice.
Loading comments…