INFRASTRUCTURE LEARNINGINFRASTRUCTURE LEARNING
インフラ大全インフラ大全IT INFRASTRUCTURE / LEARNING
Infrastructure learning guide
Find a lesson or lab/

Esc to close · Search terms stay on your device

LEARNING日本語

Topic lessonTopic: Bedrock inference, retrieval and operating decisions

Estimate Amazon Bedrock cost from input, output and add-ons

Charging depends on model and usage mode. For token pricing, record input and output separately and identify add-on terms.

What you will learn

Make a repeatable estimate for one hundred synthetic requests.

Prerequisite

The cases use fictional inputs.

What to predict

Charging depends on model and usage mode. For token pricing, record input and output separately and identify add-on terms.

From conditions to the decision
From conditions to the decision1: Calls — Align conditions — Usage. 2: Usage — Compare outcomes — Charge terms. 3: Charge terms — Verify the decision — Billing123CallsUsageCharge termsBillingFrom conditions to the decision1: Calls — Align conditions — Usage. 2: Usage — Compare outcomes — Charge terms. 3: Charge terms — Verify the decision — Billing123CallsUsageCharge termsBilling
  1. Calls → UsageAlign conditions
  2. Usage → Charge termsCompare outcomes
  3. Charge terms → BillingVerify the decision
Calls
100 calls at 200 input tokens
Usage
100 calls at 80 output tokens
Charge terms
Retrieval and guardrails are in scope
Billing
Shorter output omits required facts

Compare the cases

Synthetic observations
CaseConditionsDecision
input volume100 calls at 200 input tokensCount 20,000 input tokens
output volume100 calls at 80 output tokensCount 8,000 output tokens
ancillary usageRetrieval and guardrails are in scopeCheck applicable separate charge terms
qualityShorter output omits required factsCompare cost and required quality

Acceptance and limits

Figures are illustrative. Check official rates, caching, retrieval and guardrails by usage mode.

Measure calls and token directions.

Check ancillary fees.

Compare options that retain required facts.

Check your understanding

Choose an option to see the result and explanation.

Which decision applies to input volume?
Read the answer and reason

Answer: Count 20,000 input tokens · input volume: 100 calls at 200 input tokens → Count 20,000 input tokens

Which decision applies to output volume?
Read the answer and reason

Answer: Count 8,000 output tokens · output volume: 100 calls at 80 output tokens → Count 8,000 output tokens

Which decision applies to ancillary usage?
Read the answer and reason

Answer: Check applicable separate charge terms · ancillary usage: Retrieval and guardrails are in scope → Check applicable separate charge terms

Check, then explain in your own words

  1. Measure calls and token directions.
  2. Check ancillary fees.
  3. Compare options that retain required facts.
Check your understanding: Make a repeatable estimate for one hundred synthetic requests.

Figures are illustrative. Check official rates, caching, retrieval and guardrails by usage mode.

Compare options that retain required facts.

Sources and conditions checked

The official references used in this lesson are collected here with their scope, conditions and corresponding sections. Independent expert review and live AWS comparison have not been performed.