- Schema:
CORTEX_DOCUMENT_PROCESSING_USAGE_HISTORY view¶
This Account Usage view displays Document AI processing function activity, including Document AI’s PREDICT and PARSE_DOCUMENT calls. It shows pages processed and credits used, aggregated hourly by function and model. The view includes metadata such as the following:
Warehouse ID
Execution timestamps
Function names
Model names
Columns¶
Column Name |
Data Type |
Description |
---|---|---|
CREDITS_USED |
NUMBER(38,9) |
The number of credits billed for Cortex Document processing functions for the specified query |
START_TIME |
TIMESTAMP_LTZ |
Start of the query |
END_TIME |
TIMESTAMP_LTZ |
End of the query |
FUNCTION_NAME |
TEXT |
The name of the Cortex Document processing function |
MODEL_NAME |
TEXT |
The name of the model |
OPERATION_NAME |
TEXT |
|
PAGE_COUNT |
NUMBER |
The number of pages processed |
DOCUMENT_COUNT |
NUMBER |
The number of documents processed |
FEATURE_COUNT |
NUMBER |
The number of data values defined for document processing operations that involve entry extraction |
Usage notes¶
The view provides up-to-date credit usage for an account within the last 365 days (1 year).
Credit rate usage is based on the number of messages processed, as outlined in the Snowflake Service Consumption Table.