Option A is incorrect. The
VECTOR
data type is not supported as a clustering key. Option B is incorrect. The VECTOR data type is not supported for use with dynamic tables. Option C is incorrect. Snowflake recommends executing queries that call Cortex AI SQL functions with a smaller warehouse (no larger than MEDIUM), as larger warehouses do not increase performance. This guidance applies to functions like embedding generation, and vector similarity functions do not incur token-based costs, so performance scaling based on warehouse size for the function itself is not a factor in the same way. Snowpark-optimized warehouses are typically recommended for workloads with large memory requirements or specific CPU architectures, not general Cortex AI function calls. Option D is correct.
VECTOR_L1_DISTANCE
is a native SQL function and can be used directly in SQL queries, which are the core component of Snowflake tasks for automating data pipelines. Option E is incorrect. The VECTOR data type and vector similarity functions are supported in SQL, not exclusively in Python UDFs.