Amazon Bedrock · via your AWS connection
LiveAmazon Bedrock cost tracking, by model and token type
On your AWS bill, Bedrock is one service line. CloudQuell splits it by model, token type, and routing, and puts it beside your direct OpenAI and Anthropic spend.
It isn't a separate connector: Bedrock arrives through your AWS connection, from the Cost and Usage Report you already share, and shows up in the AI Spend view on its own once your bill carries Bedrock charges.
What CloudQuell reads
Bedrock's line items from your AWS Cost and Usage Report — the same read-only export behind the rest of your AWS view. Each line's usage type says which model ran and what kind of tokens it billed.
- Model
- Parsed from the usage type, so Claude on Bedrock reads as "Claude Sonnet 4.6", not a raw meter code — with spend by model, a model cost comparison table, and spend over time stacked by model.
- Token type
- Input (uncached), output, cache read, and cache write, each priced separately — plus a non-token bucket for provisioned capacity, guardrails, and tools, so it never skews the per-token math.
- Routing
- In-region versus cross-region (global) inference, read from the same usage type.
- Account, region, tag
- Bedrock spend by linked account, by AWS region, and by any cost-allocation tag key you pick.
- History
- On connect, the 12-month Cost Explorer backfill brings in Bedrock's dollars by account. Model and token detail come from the CUR and build from there.
What the data looks like in AI Spend
One Bedrock line item, decoded. The usage type carries the model, the token type, and the routing; CloudQuell keeps all three.
| AI Spend field | CUR source | Example |
|---|---|---|
| Source | product code AmazonBedrock | Amazon Bedrock |
| Model | usage type | Claude Sonnet 4.6 |
| Token type | usage type | Output |
| Routing | usage type | Cross-region (global) |
| Linked account | line-item account | 4021…8843 (prod) |
| Region | line-item region | us-east-1 |
| Cost (USD) | line-item cost, per your AWS cost settings | $318.75 |
How it connects: via your AWS connection
There is no Bedrock connector. If CloudQuell can read your AWS bill, it can read your Bedrock spend.
- Connector
- Your AWS integration — the read-only cross-account IAM role and CUR export. Nothing extra to install, grant, or key in.
- When it appears
- A Bedrock view turns on in AI Spend by itself once your AWS bill carries Bedrock charges, usually within a day of them landing.
- Counted once
- Bedrock stays inside your AWS cloud spend. AI Spend is a read-only lens over those same rows, and its All AI total is never added to cloud totals — so the Overview, budgets, and allocation still reconcile to your invoices.
- Cost basis
- Bedrock follows the cost settings on your AWS integration, the same as every other AWS line.
- Plans
- AI Spend, Bedrock included, is on every plan — Free covers up to $10K of monthly cloud spend. Alerts and budgets are on paid plans.
What it unlocks
Bedrock gets the same treatment as the rest of your AI bill — without being counted twice.
All your AI spend, side by side
Switch AI Spend between All AI, Anthropic, OpenAI, and Bedrock — plus Azure AI Foundry and Gemini / Vertex AI usage from your Azure and Google Cloud bills, where you have it. Under All AI, Claude on Bedrock and Claude direct share one model line.
Anthropic integration →Effective $ per 1M tokens
A blended rate over token-metered spend only, with a note on how much of your Bedrock bill it covers — plus cache-read share and a projected month-end.
Anomaly alerts
Bedrock is watched as a service on your AWS bill: when it breaks from its own baseline it is flagged, ranked by dollar severity, and can alert Slack, Microsoft Teams, or email.
AWS cost anomaly detection →Budgets and allocation
Budget the AmazonBedrock service or a tag, and route Bedrock to cost centers with tag or service rules — the same mechanics as any other AWS cost.
Where Bedrock fits in an AI bill
Bedrock is one of three places AI spend lands: direct API invoices, managed models on the cloud bill, and GPU infrastructure. Our guide to AI cost management tools covers which tools see which layer — including the free native options on AWS.
AI cost management tools compared →Common questions
- Do I need a separate integration to track Bedrock costs?
- No. Bedrock charges are part of your AWS bill, so they arrive through your existing AWS connection — the same read-only IAM role and Cost and Usage Report. Once those charges appear, a Bedrock view turns on in AI Spend by itself, usually within a day. There is nothing extra to install or grant.
- Is Bedrock spend double-counted with my AWS total?
- No. Bedrock is counted once, inside your AWS cloud spend. AI Spend is a read-only lens over those same rows: its All AI figure combines direct and cloud-billed AI spend for analysis and is never added to cloud totals, budgets, or allocation — so every number still reconciles to the invoices you receive.
- Can I see Claude on Bedrock costs by model?
- Yes. CloudQuell reads the model from each line's usage type, so Claude on Bedrock shows as named models such as Claude Sonnet 4.6, split into input, output, cache-read, and cache-write tokens. Under All AI it sits beside your direct Anthropic usage, which comes in through the Anthropic integration.
- Can I get alerts and budgets on Bedrock spend?
- Yes, on paid plans. Anomaly detection watches the Bedrock service line on your AWS bill against its own baseline and alerts Slack, Microsoft Teams, or email; budgets can be scoped to the AmazonBedrock service or a tag. Alerts work at the service level — the per-model breakdown lives in the AI Spend view.
- What does Bedrock cost tracking cost?
- AI Spend, Bedrock included, is on the Free plan for up to $10K of monthly cloud spend. Paid plans are a flat monthly fee — Starter $99/month up to $50K of monthly spend, Growth $199/month up to $200K — with a 14-day trial and no credit card. Never a percentage of your bill.
Free under $10K/month of cloud spend. Bedrock comes in with your AWS connection — read-only, counted once, flat monthly price.