Amazon Bedrock pricing

Example 1: Customer support chatbot
An application developer creates a customer support chatbot and uses content filters to block harmful content and denied topics to filter undesirable queries and responses.

The chatbot serves 1000 user queries per hour. Each user query has an average input length of 200 characters and receives a FM response of 1500 characters.

Each user query of 200 characters correspond to 1 text unit.

Each FM response of 1,500 characters correspond to 2 text units.

Text units processed each hour = (1 + 2) * 1000 queries = 3000 text units

Total cost incurred per hour for content filters and denied topics = 3000 * ($0.15 + $0.15) / 1000 = $0.90

Pricing Example 1 (Reranking using Amazon Rerank 1.0 model)

In a given month, you make 2 million requests to Rerank API using Amazon Rerank 1.0 model – 1 million requests contain fewer than 100 documents each and hence will be charged for one request each. The remaining 1 million requests contain 120-150 documents, and hence each request will be charged for 2 requests.

Price for one request = $0.001
Total charge = 1,000,000 * $0.001 + 1,000,000*2*$0.001= $3000 

Pricing Example 2: (Structured data retrieval)

An application developer creates a support chatbot that queries structured data stored in Amazon Redshift. The developer creates a Bedrock Knowledge Base and connects to Amazon Redshift. The chatbot serves 10000 user queries per hour. Each user query will cost $0.002 per GenerateQuery API to generate SQL from user query.

Total cost incurred for generating SQL per hour = $0.002*10000 = $20.
Total cost incurred in month = $20*24*30 = $1440

Pricing Example: An application developer imports a customized Llama 3.1 type model that is 8B parameter in size with a 128K sequence length in us-east-1 region and deletes the model after 1 month. This requires 2 Custom Model Units. So, the price per minute will be $0.1570 because 2 Custom Model Units are required. The model storage costs for 2 Custom Model Units would be $3.90 for the month.

There is no charge to import the model. The first successful invocation is at 8:03 AM, at which time the metering starts. The 5-minute metering windows are from 8:03 AM - 8:07 AM; 8:07 AM - 8:11 AM, and so on. If there is at least one invocation during any 5-minute period, the window will be considered active for billing. If there is an invocation at 8:03 AM and no further invocations after 8:07 AM, the metering will stop at 8:07 AM. In this case, the bill would be calculated as follows: $0.1570 * 5 minutes * 1 five minute windows = $0.785.

Pricing example 1:
Let’s say you process a 1,000 page document using BDA Custom Output. All 1,000 pages are processed using blueprint 1 which has 15 fields. The per page price for any blueprint with 30 fields or less is $0.040. The total cost would be $40.

Total pages processed = 1,000
Price per page for blueprints with less than 30 fields = $0.040
Total charge = 1,000 * $0.040 = $40

Pricing example 2:
Let’s say you process 2 documents using BDA Custom Output. Document 1 has 40 pages and is processed using blueprint 1 which has 20 fields. Document 2 has 10 pages and is processed using blueprint 2, which has 40 fields. The per page price of blueprint 1 is $0.040 since it contains 30 fields or less. The per page price of blueprint 2 is $0.045. The processing cost for Document 1 using blueprint 1 is $1.60. The processing cost for Document 2 using blueprint 2 is $0.45. The total cost of processing both documents would be $2.05.

Total pages processed = 50
Price per page for Blueprint 1 with less than 30 fields = $0.040
Price per page for Blueprint 2 with 40 fields = $0.040 + (# of additional fields above 30 *$0.0005 per field)
Number of additional fields above 30 = 40 - 30 = 10
Price per page for Blueprint 2 with 40 fields = $0.040 + (10 *$0.0005 per field) = $0.045
Charge for Document 1 using Blueprint 1 = 40 pages x $0.040 per page = $1.6
Charge for Document 2 using Blueprint 2 = 10 pages x $0.045 per page = $0.45
Total charge = Charge for Document 1 + Charge for Document 2 = $1.6 + $0.45 = $2.05

Pricing Example 3:
Let’s say you setup Bedrock Knowledge Bases to use Bedrock Data Automation as a parser and then ingest a 1,000 page document. Note, cost structures differ between the Knowledge Bases parsing options. BDA uses per-page pricing, while Foundational Model parsers charge based on input and output tokens. For context, processing 1,000 pages, where 30% contain tables and 30% contain figures, typically requires 2,900 input tokens and 750 output tokens. Token consumption varies by content type, so customers are encouraged to test using their own data to get more accurate estimates. Bedrock Knowledge Bases and Bedrock Data Automation integration uses standard output, where the per page price is $0.010. The total cost would be $10.

Total pages processed = 1,000
Price per page for standard output = $0.010
Total charge = 1,000 * $0.010 = $10

Pricing example 4:
Let’s say you process a 60 minute video using BDA Standard Output. The per minute price for video standard output is $0.050. The total cost would be $3.00.

Total minutes processed = 60
Price per minute for video standard output = $0.050
Total charge = 60 * $0.050 = $3.00

Pricing example 5:
Let’s say you process 2,000 images using BDA Custom Output. The first 1,000 images are processed using blueprint 1, which has 10 fields. The last 1,000 pages are processed using blueprint 2, which has 40 fields. The per image price for blueprint 1 is $0.005, since it contains 30 fields or less. The per image price of blueprint 2 is $0.01. The processing cost for the first 1,000 images using blueprint 1 is $5.00. The processing cost for the second 1,000 images using blueprint 2 is $10.00. The total cost of processing all 2,000 images would be $15.00

Cost for first 1000 images = 1,000 images * $0.005 per image = $5.00
Cost for second 1,000 images = 1,000 images * ($0.005 + (# of additional fields above 30 *$0.0005 per field))
= 1,000 * ($0.005 + ((40-30)*$0.0005))
= 1,000 * ($0.005 + (10*$0.0005)) = $10.00
Total cost = $5.00 + $10.00 = $15.00

Pricing example 6:
Let’s assume that you want to use Bedrock Data Automation Standard Output to process 15,000 minutes of meeting audio recordings in your organization. The total cost of processing all 15,000 audio minutes would be $90.

Total minutes processed = 15,000 minutes
Total charge = 15,000 min × $0.006 = $90

On-Demand pricing

An application developer makes the following API calls to Amazon Bedrock on an hourly basis: a request to the DeepSeek-R1 model to summarize an input of 2K tokens of input text to an output of 1K tokens (including reasoning tokens):

Total hourly cost incurred = 2K tokens/1000 * $0.00135 + 1K tokens/1000 * $0.0054 = $0.0081

Example: News summarization
An application developer creates a flow to automate news summarization for traders. The flow includes an Input node that takes in an S3 location, and a S3 retrieval node that retrieves 10 files that include articles from 10 major news agency in S3 (2 node transitions). It then uses an iterator node to invoke a model with a prompt node to summarize each file (+ 10 files x 2 node transitions). It then collects all the results using a collector node, write the results to S3 using S3 storage node, and complete in an Output node (+ 3 node transition). They run this flow every half hour of every week day.

The number of node transition per flow execution is: 2+1+10*2 + 3 = 25 node transitions/flow execution

The number of flow execution per month is: 24 hours *2* 5 days * 4 weeks = 960 flow executions/month.

Total monthly bill is: 25 * 960 * $0.035/1000 = $0.84

Additional charges
The bill will also include additional charges for AWS services used in the workflow execution, including Amazon S3 usages in the retrieval and storage nodes, and Amazon Bedrock foundation model usage in the prompt node.