Vector & Embedding Similarity Studio

Compare embedding vectors with cosine similarity, dot product, euclidean and manhattan distance metrics

Embedding & Vector Presets

Load sample embedding vectors across different semantic distributions

Vector A (Source Embedding)

Accepts JSON array [0.1, 0.2] or comma/space-separated numbers

8D (||A|| = 1.31758)

Normalized: [0.18595, 0.61628, -0.31269, ...]

Vector B (Target Embedding)

Accepts JSON array [0.1, 0.2] or comma/space-separated numbers

8D (||B|| = 1.33443)

Normalized: [0.16337, 0.63323, -0.29151, ...]

Directional alignment between vectors independent of magnitude

Very High Semantic Match

0.99825[-1.0 ↔ +1.0]

Match Affinity: 100%

-1.0 (Opposite)0.0 (Orthogonal)+1.0 (Exact)

1.75515

Sum of products; combines magnitude and directional alignment

0.0802

Straight-line geometric distance between vectors in N-dimensional space

0.00175

Normalized angular difference between vectors (1 - similarity)

0.224

Sum of absolute coordinate differences (grid / taxicab distance)