Embedding Cosine Similarity Calculator
Compare two embedding vectors with cosine similarity, dot product, and Euclidean distance.
Cosine Similarity
0.9988
| Dot Product | 0.9410 |
| Euclidean Distance | 0.0686 |
| Dimensions | 4 |
About Embedding Cosine Similarity Calculator
When working with text embeddings — from OpenAI, Cohere, or any embedding model — similarity between two vectors is usually measured with cosine similarity: a score from -1 (opposite meaning) to 1 (identical meaning), independent of vector magnitude. Paste two vectors (as a JSON array or space/comma-separated numbers) to compute their cosine similarity, along with the raw dot product and Euclidean distance. Useful for debugging a RAG pipeline, sanity-checking an embedding model's output, or just understanding how similarity search actually works under the hood.
Frequently Asked Questions
1 means the vectors point in exactly the same direction (maximally similar), 0 means they're orthogonal (unrelated), and -1 means they point in opposite directions. For text embeddings, scores are usually positive, with values above ~0.8 generally indicating strong semantic similarity.
Cosine similarity only considers the angle between vectors, not their magnitude, which matters because embedding magnitude often reflects things like text length rather than meaning. Most vector databases and embedding models are designed around cosine similarity for this reason.
Either a JSON array like [0.12, -0.45, 0.88] or plain numbers separated by spaces or commas — both are accepted.