@leuria/web-embed

A small embedding model that runs in the page, as a Leuria provider: search by meaning for visitors whose own AI can't embed. It uses Transformers.js in a Web Worker, so the page stays responsive.

import { leuria, promptAPI, createAI } from "@leuria/client"
import { pageEmbedder } from "@leuria/web-embed"

const ai = createAI({ providers: [leuria(), promptAPI(), pageEmbedder()] })
// Later, from a click (the visitor agrees to the download):
await ai.connect("page-embed")

• Asks first. Until the visitor agrees, the provider is needs-action with action: "download". The model and its runtime are about 40 MB, downloaded once and kept in the browser's cache. autoLoad: true skips the question.
• Default model: Xenova/bge-small-en-v1.5 (English, made for search), with the query instruction from its model card. Choose another with model (and queryPrefix).
• Where files come from: Hugging Face by default; set remoteHost to serve the model from your own site. Texts never leave the page.

Apache-2.0.