Home icon

Build an internal SaaS service with cost and usage tracking for foundation models on Amazon Bedrock

Machine Learning Blog



This article provides a technical guide on building an internal SaaS platform for accessing foundation models on Amazon Bedrock, with a focus on cost and usage tracking for different teams or tenants.

Specifically, the article covers:

  • Solution architecture and key components for capturing model usage and tracking costs per team
  • Controlling usage limits and throttling requests for each team using API Gateway usage plans
  • Configuring access control to specific models for each team
  • Deploying the solution using an AWS CDK template
  • Invoking the service with code examples for text generation and embeddings
  • Cost estimation example based on input/output token usage
  • End-to-end view of a multi-tenant serverless SaaS environment
  • Support for different Amazon Bedrock consumption plans (on-demand and provisioned throughput)


Go to article

The AWS News Feed is currently looking for gold sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.

Related articles

Apr 22
2025
Optimizing cost for using foundational models with Amazon Bedrock
Nov 6
2024
Integrate foundation models into your code with Amazon Bedrock
Aug 4
2025
Cost tracking multi-tenant model inference on Amazon Bedrock
Apr 7
2026
Manage AI costs with Amazon Bedrock Projects

The AWS News Feed is currently looking for silver sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.