AWS cost tools add model-level Amazon Bedrock product attributes
Finance and platform teams can see which models and workloads drive Bedrock costs within their existing AWS cost tools.
What changed
Source summary
On October 8, 2026, AWS added Amazon Bedrock product attributes to Cost Explorer, Budgets and Cost Management Dashboards. Teams can analyse spending by model, provider, inference type and feature. Cost Explorer and Dashboards support grouping and filtering; Budgets supports filtering. These attributes can be combined with allocation tags for applications and IAM principal tags for users or roles. AWS says the dimension is available without an additional charge in all regions except AWS GovCloud (US) and China Beijing and Ningxia.
Why it matters to your work
Practical interpretation
Finance and platform teams can see which models and workloads drive Bedrock costs within their existing AWS cost tools.
When this could help
Useful when you need to identify which AI models or workloads drive the most Bedrock spend, set per-model budget alerts, or build dashboards tracking cost per provider.
AWS Cost Explorer, Budgets, and Cost Management Dashboards now support Bedrock-specific product attributes, letting you group and filter spend by individual model, model provider, inference type (input/output tokens), and feature (on-demand inference, reranker, etc.) — previously Bedrock appeared as a single line item.
How to get started
Open AWS Cost Explorer and group or filter by the new Bedrock product attributes (model, model provider, inference type, feature).
Provider guidance ↗Pair attributes with cost allocation tags on application inference profiles, projects, or workspaces to see which applications drive spend per model.
Provider guidance ↗In AWS Budgets, create or update a budget filtered by a specific model to get alerts when that model's spend exceeds a threshold.
Provider guidance ↗
Build it into your routine
SCALA suggested workflow
Monthly Bedrock cost review
Group monthly Bedrock spend by model in Cost Explorer to identify the most expensive models, then review whether cheaper alternatives could handle some of those workloads.
What to check
- Not available in AWS GovCloud (US), China (Beijing), or China (Ningxia) Regions. Source ↗
Sources & provider guides
Official documentation checked 12 Oct 2026. SCALA prompts and workflows are suggestions, not claims of hands-on testing.
Source & availability
This is a summary of a published provider announcement, with practical context from SCALA’s feed.
Plan, region, workspace and rollout conditions may affect access in Australia. Check the original source for current availability.
aws.amazon.com