Unstructured augmentation credits

Unstructured augmentation processes pages from uploaded documents and configured document sources to:

• Extract information from documents and map it to entities and attributes in your Reltio tenant.

• Create or update searchable knowledge bases for custom agents.

Credits consumed per page processed

Credits are consumed according to the number of pages processed across Unstructured augmentation workflows, including:

• One Time Upload

• Automated Bulk Extraction

• Knowledge base indexing

Unstructured extraction processes pages to extract structured information and map it to entities and attributes in your Reltio tenant.

Unstructured search processes pages to create a searchable index that autonomous agents can use when answering questions.

Agent and document-processing credits

Unstructured augmentation page processing and autonomous agents conversations represent different types of usage.

Processing documents uses the Unstructured augmentation page-based credit model. When users interact with a published agent, the interaction follows the standard autonomous agents credit model.

For more information, see autonomous agents billing and credit usage.

Credit entitlement and monitoring

Unstructured augmentation credits are deducted from your organization's Autonomous agents credit entitlement. Credits are allocated for a defined contract period and do not reset monthly. Unused credits expire at the end of the contract period.

Users assigned ROLE_AGENT_BILLING can use the Usage Dashboard to monitor credit balances and consumption.

Notifications for Autonomous agents appear at the following thresholds:

• 75% — Warning notification

• 90% — Critical warning notification

• 100% — Usage is blocked until credits are replenished