Vector similarity functions

The measurement of similarity between vectors is a fundamental operation in semantic comparison. For example, you need this operation to find the top N number of closest vectors to a query vector, which can be used for a semantic search. Vector search also enables developers to improve the accuracy of their generative AI responses by providing related documents to a large language model.

Snowflake Cortex provides three vector similarity functions:

  • VECTOR_INNER_PRODUCT

  • VECTOR_L2_DISTANCE

  • VECTOR_COSINE_SIMILARITY

Each function takes two VECTOR arguments of equal element type and dimension and computes the specified metric over them.

Note

Due to computational optimizations in these functions, floating-point errors may be slightly larger than usual (e.g. about 1e-4).

List of functions

Function Name

Notes
VECTOR_INNER_PRODUCT
VECTOR_L2_DISTANCE
VECTOR_COSINE_SIMILARITY

Not supported in Snowpark API.
Language: English