Embedding Vector Inspector

Compare two embedding vectors with cosine similarity, dot product, Euclidean distance, magnitudes, and normalization.

Vector A

Vector B

Dimensions

Cosine similarity

0.994738

Dot product

1.120500

Euclidean distance

0.112250

Magnitudes A / B

1.075081 / 1.047760

L2-normalized vectors

{
  "a": [
    0.11161945585058808,
    -0.03720648528352936,
    0.8185426762376459,
    0.28835026094735255,
    -0.20463566905941147,
    0.43717620208146996
  ],
  "b": [
    0.09544174844165572,
    -0.07635339875332457,
    0.7730781623774113,
    0.334046119545795,
    -0.1717951471949803,
    0.49629709189660975
  ]
}

Private by default: this tool runs entirely in your browser. Nothing you paste is uploaded to Execute.Online.

About Embedding Vector Cosine Similarity Calculator

Compare embedding vectors with cosine similarity, dot product, Euclidean distance, magnitudes, dimension checks, and L2 normalization.

How to use Embedding Inspector

1Paste two vectors as JSON arrays or comma-separated numbers.
2Confirm both vectors have matching dimensions.
3Compare cosine, dot-product, distance, and magnitude metrics.
4Copy the normalized vectors when needed.

Frequently Asked Questions

What does cosine similarity measure?

Cosine similarity measures the angle between vectors. Values closer to 1 indicate similar directions, independent of vector magnitude.

Why must embedding dimensions match?

Vector operations compare corresponding coordinates, so both embeddings must come from compatible models and have identical dimensions.

What is L2 normalization?

L2 normalization scales a vector to a magnitude of one, which makes its dot product equivalent to cosine similarity.

Related Tools

RAG Chunker

Split documents into retrieval-ready chunks.

AI JSONL Validator

Validate chat and fine-tuning datasets.

Number Base & IEEE 754

Base converter and float bit visualizer.

Token Estimator

Estimate token counts and costs.