Distance Metrics

NeuronDB supports multiple distance metrics for measuring vector similarity.

L2 (Euclidean) Distance

Straight-line distance between two points. Lower values indicate more similar vectors.

Copy
-- L2 distance operator
SELECT embedding <-> '[0.1, 0.2, 0.3]'::vector AS distance
FROM documents
ORDER BY distance
LIMIT 10;

Cosine Distance

Measures the angle between vectors. Ideal for normalized vectors where direction matters more than magnitude.

Copy
-- Cosine distance operator
SELECT embedding <=> '[0.1, 0.2, 0.3]'::vector AS distance
FROM documents
ORDER BY distance
LIMIT 10;

Inner Product

Dot product of two vectors. Higher values indicate greater similarity. Best used with normalized vectors.

Copy
-- Inner product (negative for ordering)
SELECT embedding <#> '[0.1, 0.2, 0.3]'::vector AS neg_inner_product
FROM documents
ORDER BY neg_inner_product
LIMIT 10;

Other Metrics

NeuronDB also supports:

• Manhattan (L1) distance
• Hamming distance (for binary vectors)
• Jaccard distance (for sets)

Learn More

For detailed documentation on all distance metrics, when to use each, GPU acceleration, and performance characteristics, visit:

Distance Metrics Documentation

Related Topics

• Vector Types - Vector data types
• Indexing - Creating indexes for fast search