Skip to content

[Feature] Add Application Inference Profile support for Bedrock cost allocation #151

Description

@juengler

Description:

The solution does not support Application Inference Profiles, which are the AWS-recommended mechanism for cost allocation and usage tracking on Bedrock. Application Inference Profiles allow customers to attach tags to on-demand model invocation requests for cost tracking.

Expected behavior:

  • Users should be able to configure an Application Inference Profile ARN in their profile config.
  • When configured, the profile ARN should be used as the modelId in Bedrock API calls instead of the CRIS model ID.

Impact: Customers cannot use AWS-recommended cost allocation tagging for Bedrock usage without modifying the Python scripts themselves.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions