Categories:

String & binary functions (Large Language Model)

EMBED_TEXT_1024 (SNOWFLAKE.CORTEX)¶

Creates a vector embedding of 1024 dimensions from text.

Syntax¶

SNOWFLAKE.CORTEX.EMBED_TEXT_1024( <model>, <text> )
Copy

Arguments¶

model

A string specifying the vector embedding model to be used to generate the embedding. This must be one of the following values.

  • nv-embed-qa-4 (English only)

  • multilingual-e5-large

  • voyage-multilingual-2

Supported models might have different costs.

text

The text for which an embedding should be calculated.

Returns¶

A vector embedding of type VECTOR.

Access control requirements¶

You must use a role that has been granted the SNOWFLAKE.CORTEX_USER database role. See Required privileges for more information on granting this privilege.

Example¶

In this example, a vector embedding is generated for the phrase hello world using the nv-embed-qa-4 model:

SELECT SNOWFLAKE.CORTEX.EMBED_TEXT_1024('nv-embed-qa-4', 'hello world');
Copy

In this example, a vector embedding is generated for the Spanish phrase hola mundo using the multilingual-e5-large model:

SELECT SNOWFLAKE.CORTEX.EMBED_TEXT_1024('multilingual-e5-large', 'hola mundo');
Copy