Start a project

Tool 46 of 425 · Embeddings

Embedding Similarity Calculator

Compares embedding vectors pairwise with the common similarity measures.

Embedding Similarity Calculator

Your input

Result

PairCosineDotEuclidean

    01

    What it does

    Compares embedding vectors pairwise with the common similarity measures.

    • Cosine, dot product, Euclidean
    • Labelled vectors
    • Dimension mismatch check
    • Normalisation check

    02

    Private by design

    Everything happens in your browser. Nothing you type, paste or upload is sent anywhere, and it keeps working offline once the page has loaded.

    03

    What you get

    A similarity table, with a Copy and a Download button on every result.

    • Cosine similarity
    • Embeddings
    • Vectors

    Why it matters

    Debugging semantic search means checking whether two embeddings really are close.

    Paste vectors from your model's output and see every pair ranked.

    How to use it

    1. Type or paste your input; the example shows the format.
    2. Press Compare. Results also update as you type.
    3. Read the summary line, then the table, notes or output below it.
    4. Copy or download what you need. Nothing leaves this page.

    FAQ

    Cosine or dot product?

    For normalised vectors they rank identically; cosine is safer otherwise.

    What's a good cosine score?

    It depends on the model; compare against known related and unrelated pairs.

    Is my input uploaded?

    No. The tool runs in your browser and nothing you type, paste or open is sent anywhere.

    Can I rely on the result?

    It follows the public specification or formula named on this page and says where it simplifies. Check anything critical against the official tool or documentation before you ship it.

    Work with me

    Want a site that gets these right?

    I build fast, accessible websites and web apps that pass these checks from day one. Tell me about your project and I'll reply personally with honest next steps.

    All developer tools · Back to leej.uk