Cost attribution and controls for AI developer and productivity tools on Amazon Bedrock
Amazon Bedrock provides built-in, zero-instrumentation cost attribution that traces every inference call back to the IAM principal that made it. This post shows how to attribute per-user Bedrock costs for AI developer and productivity tools using IAM Identity Center, session tags, and CUR 2.0, and how to enforce per-user usage limits.
First seen by Oriel on .
AI summary
This AWS guide explains how Amazon Bedrock provides per-user cost attribution and organizational controls for AI coding and productivity tools through IAM Identity Center, CUR 2.0, Athena, and billing dashboards.
- Bedrock records each inference caller’s IAM ARN at the bedrock-runtime or bedrock-mantle endpoint, without changes to tool API calls or custom instrumentation.
- IAM Identity Center federated sessions and session tags such as Team and CostCenter make user and organizational spending queryable in CUR 2.0 and Cost Explorer.
- CUR 2.0 aggregates costs by principal and usage type per export interval, while model invocation logs or request metadata provide finer per-prompt detail.
- The setup covers Claude Code, Claude Desktop, OpenAI Codex, ChatGPT Work, and OpenCode, with endpoint-specific permissions and workstation configuration.
- AWS recommends Athena queries and dashboards for analysis, but cost-allocation tags are not retroactive and may take up to 48 hours to appear in CUR 2.0 and Cost Explorer.
#amazon bedrock #cost attribution #iam identity center #finops
Generated by gpt-5.6-sol. Check the original for details.
Aliases
| Source | External ID | URL |
|---|---|---|
| AWS Machine Learning Blog | 94b792ec6f46dfd34db3ccc1fda8cfa7fee6df93 | https://aws.amazon.com/blogs/machine-learning/cost-attribution-and-controls-for-ai-developer-and-productivity-tools-on-amazon-bedrock/ Read discussion |