Standard for websites & extensions to use local models, proposed by Opera
Summary
Opera has proposed a new standard enabling websites and browser extensions to run and leverage local AI models, aiming to standardize on-device AI capabilities across the web ecosystem.
Similar Articles
Opera introduces a new web standard to let users deploy AI locally in their browser
Opera introduced WebAI, a new web standard that lets users run open-weight AI models like Google's Gemma and Apple's on-device foundation model directly in the browser without cloud reliance, debuting as an early preview in Opera One.
WebAI by Opera
Opera launches WebAI, a feature that lets websites run AI models locally on the user's computer through the browser.
@DanKornas: Local AI models are often disconnected from the webpages you need to read, question, and work with. Page Assist is an o…
Page Assist is an open-source browser extension that adds a sidebar and web UI for chatting with local AI models on any webpage, supporting Ollama, Gemini Nano, and OpenAI-compatible endpoints.
Automatically detecting AI text in my browser (5 minute read)
The article details the creation of Deckard, a Chrome extension that leverages local AI models to automatically detect AI-generated text in web browsers, including benchmark comparisons of various models for effectiveness and privacy.
I made a Chrome extension that filters your YouTube feed with a small local model running in the browser (WebGPU, no server)
A Chrome extension that filters YouTube feeds using a small on-device ONNX decision model (opendecider-nano) running via WebGPU/WASM with no server or API calls. On 400 videos it achieves 0.934 balanced accuracy on rules like "hide music" or "only news", and can also flag prompt injection in selected text.