Opera launches WebAI for on-device AI inside websites

Opera has introduced WebAI, a browser feature that lets websites and extensions use AI models running on a user’s own computer. The launch matters because it pushes AI features closer to the web apps people already use, while Opera says page content does not need to be sent to a cloud AI service.

The feature is launching on Product Hunt today and is described there as a free, opt-in early preview in Opera One. For creators, marketers, freelancers and small teams, the bigger question is whether useful AI tasks can move from separate chatbots and add-ons into the browser itself.

What Opera says WebAI does

According to the Product Hunt launch page, WebAI gives sites and extensions access to local AI built into the browser. Opera’s pitch is that developers can build browser capabilities directly instead of depending on other people’s extensions.

A launch post on the page says Opera has placed the inference engine inside the browser. That means websites and extensions can call AI models that run locally on the device, rather than routing page content to a cloud AI provider.

Opera is exposing three APIs for developers:

• text generation
• classification
• finding related content

Those categories point to practical, lightweight web tasks: drafting short text, sorting content, or matching parts of a page to a user’s current work. Product Hunt also lists the launch tags as API, privacy and artificial intelligence, underlining that this is aimed as much at developers as at everyday browser users.

Early examples: article quizzes and relevance tools

Opera has also created six example extensions, according to the Product Hunt post. Two examples are described: one can quiz a reader on an article, and another can highlight paragraphs that are relevant to a project.

Those examples are small, but they help explain why on-device browser AI is interesting. A student, researcher, marketer or solo operator often works across web pages, notes and reference material. If a browser can classify or summarize local page context through approved APIs, developers may be able to build narrower tools that sit closer to the reading and research workflow.

This also fits a broader pattern of browsers becoming AI work surfaces, not just places to open web apps. Mozilla’s recently announced Firefox Smart Window, for example, also points toward AI features that understand browser context. OpenAI has similarly been moving toward browser context with a ChatGPT Chrome extension that can use open tabs. Opera’s angle, based on the Product Hunt description, is local model access through Opera One.

Privacy is the clearest near-term pitch

The most concrete benefit Opera is emphasizing is privacy. The Product Hunt post says websites and extensions can use models on the computer “without sending page content” to a cloud AI service.

That distinction could matter for small businesses and freelancers who handle client drafts, research notes, internal planning pages or private source material in the browser. Even when cloud AI tools are useful, many teams still need to think carefully about what page content or client information leaves the device.

Opera has not, in the Product Hunt text, provided performance comparisons, model sizes, enterprise controls or a detailed security model for WebAI. So the practical takeaway is narrower: WebAI is an early preview of browser-level local AI access, not yet a fully documented replacement for hosted AI services.

The page says WebAI supports local models including Gemma and Apple’s on-device foundation model on Macs. That suggests Opera is not presenting WebAI as a single-model feature, though the source does not list every supported model or platform.

What creators and small teams should watch

Because WebAI is described as an opt-in early preview in Opera One, the first audience is likely developers and technically curious users rather than teams looking for a finished workflow today.

For creators, the useful test will be whether WebAI extensions can solve focused jobs better than general AI chat interfaces: checking whether an article covered key points, pulling out relevant passages, classifying research pages, or generating small snippets without leaving the browser.

For small businesses, the question is governance. Local processing may reduce the need to send page content to external AI services, according to Opera’s description, but teams will still need to understand which websites and extensions are allowed to invoke the APIs and what permissions are involved. The Product Hunt listing does not spell out those controls.

Opera already has a history of AI-related browser launches on Product Hunt, including Opera Aria for iOS and Opera Operator. WebAI is different in emphasis: rather than presenting one built-in assistant, it gives developers APIs to build AI behavior into websites and extensions that run through the browser.

If that developer layer becomes useful, the browser could become a more important place for everyday AI work. For now, WebAI is best viewed as an early preview worth watching by anyone building or using web-based productivity tools.