Track generative AI costs with Amazon Bedrock inference profiles

This article explains how to track generative AI costs by department using Amazon Bedrock application inference profiles and cost allocation tags.

• Create tagged inference profiles for each department pointing to the same foundation model
• Activate cost allocation tags in AWS Billing and Cost Management console
• Route application requests to department-specific inference profile ARNs based on user authentication
• View per-department cost breakdown in Cost Explorer grouped by Team tag
• No additional charges for cost attribution; pay standard per-token rates
• Extend cost tracking to Knowledge Bases and other Bedrock features using same tagged profiles

This approach enables finance teams to chargeback departments, set per-team budgets, and identify spending drivers without requiring separate IAM identities per team.