- Schema:
CORTEX_ DOCUMENT_ PROCESSING_ USAGE_ HISTORY view¶
This Account Usage view displays document processing function activity, including PARSE_DOCUMENT (SNOWFLAKE.CORTEX),
AI_EXTRACT, and <model_build_name>!PREDICT 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 |
|---|---|---|
| QUERY_ID | VARCHAR | A unique identifier for the SQL query |
| 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 hourly time range in which the query usage took place. |
| END_TIME | TIMESTAMP_LTZ | End of the hourly time range in which the query usage took place. |
| 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.