Skip to content
runonweb

Text embeddings

Vector

Turn sentences into 384-dimensional vectors for semantic search, clustering or deduplication. Compare meaning, not keywords, in the browser, in milliseconds.

Try it nowDocs

Try Vector

Runs in this tab · nothing uploaded

Pick a sample or type your own, then rank
Try with

    What it does

    Semantic search in 23 MB.

    • Vectors are L2-normalized by default, so cosine similarity is a dot product.
    • Works best on English; pass a multilingual model via model if needed.

    Specifications

    Package
    runonweb/embed
    Base model
    all-MiniLM-L6-v2 · sentence-transformers
    License
    Apache-2.0
    Weights
    Xenova/all-MiniLM-L6-v2
    Download
    ~45 MB / ~23 MB
    Quantization
    fp16 on WebGPUq8 on WASM
    Backend
    WebGPUWASM fallback
    Input
    Text

    Get started with runonweb/embed

    pnpm add runonweb
    runonweb/embedts
    import { TextEmbedder, cosineSimilarity } from 'runonweb/embed'

const embedder = new TextEmbedder()
await embedder.load()

const { embeddings } = await embedder.embed([
  'How do I reset my password?',
  'I forgot my login credentials',
  'What is the weather today?',
])

cosineSimilarity(embeddings[0], embeddings[1]) // ~0.7
cosineSimilarity(embeddings[0], embeddings[2]) // ~0.0

embedder.dispose()

    Same shape everywhere: construct, load(), run, dispose(). Pass onProgressto show download progress on first use.

    Questions about Vector

    Does Vector run on device?

    Yes. Weights download once and stay in the browser. Nothing is sent to a server.

    What does Vector cost?

    Nothing. runonweb is free and MIT-licensed, with no device limits, tokens or sign-in.

    Which browsers support Vector?

    Chromium browsers use WebGPU. Firefox and Safari fall back to WebAssembly automatically, slower but identical output.

    Can I use a different model with runonweb/embed?

    Yes. See each module’s docs for the options it accepts (model, size, language…). Check the license and test both backends before shipping.