Post

Firefox for Web Developers

@webdevs.firefox.com

Chrome looks set to ship an LLM Prompt API to the web platform. At Mozilla, we oppose this API. We feel it has a large interoperability risk, and Google imposing T&Cs on a web API sets a dangerous precedent. Full details: github.com/mozilla/stan...

github.comPrompt API · Issue #1213 · mozilla/standards-positionsSpecification title Prompt API Specification or proposal URL (if available) No response Explainer URL (if available) https://github.com/webmachinelearning/prompt-api/blob/main/README.md Proposal au...

Andreas Møller

@andreasmoller.dk

... and we already have the fetch API, which solves this very well today.

Exactly what I was going to say. LLMs are just one more service you might want to call. And just like any external service, you need exercise extreme caution in what you let it do to your JavaScript and DOM.

The difference here between fetch and the prompt API is that the Prompt API is running client side with a pre-fetched model. Big difference.

Andreas Møller

@andreasmoller.dk

Can’t we already do that today? Is this just to prevent loading a library?

Firefox for Web Developers

@webdevs.firefox.com

learn.microsoft.com/en-us/window... Microsoft did look into lower level ways of making local models work. It's not without its own issues though.

learn.microsoft.comWebNN OverviewThe Web Neural Network API (WebNN) is an emerging web standard that allows web apps and frameworks to accelerate deep neural networks with on-device hardware.

30/4/26, 2:42 pm0 reposts2 likes

@alexsuzuki.com

what happened to WebNN? I seemed like a promising way to access on-device ML from web apps.

Ingvar Stepanyan

It's still worked on, pretty fast in recent builds of onnxruntime.

WebNN is for sure part of the answer here too. Different trade offs (e.g. you have to download a GB model), but part of the answer too, I think.