Minimal serverless Q&A
The cheapest starting point: API Gateway + Lambda + Bedrock, no vector store.
UserAPI GatewayLambdaBedrock (context in prompt)
Example estimate
Use this when your knowledge fits in the prompt (a handful of docs). No OpenSearch floor, no ingestion — just pay per question. Great for MVPs and internal tools.
$137.01/month
1,000 questions/day · 4,000 context tokens/question
Estimate your workload
Model choice is usually the biggest cost lever. Nova Micro/Lite are ~10–70× cheaper than Premier.
Results update automatically as you type.
Result
$137.01/month
1,000 questions/day · 4,000 context tokens/question
- Bedrock (Amazon Nova Pro)(context stuffed in prompt)
- $136.80
- Lambda
- $0.1810
- API Gateway
- $0.0300
- Per month(all services)
- $137.01
- Per year
- $1,644.13
| Bedrock (Amazon Nova Pro) | $136.80 | 100% |
| Lambda | $0.1810 | 0% |
| API Gateway | $0.0300 | 0% |
Frequently asked questions
When is 'no vector store' the right call?+
When your knowledge base is small enough to stuff into the prompt (say under ~50k tokens). Beyond that, per-question input costs rise and a RAG vector store becomes cheaper.
Other AWS architectures
Amazon Bedrock model pricing
Estimate the token cost of any Amazon Nova model on Bedrock, before you build.
Bedrock AgentCore agent cost
Predict a Bedrock AgentCore agent bill: model tokens + Runtime + Memory + optional tools.
RAG chatbot on Bedrock
Estimate a retrieval-augmented chatbot: OpenSearch vectors + Bedrock generation + Lambda.
Batch document pipeline
Estimate a high-volume document extraction/summarisation pipeline with batch pricing.
Voice agent on AWS
Estimate a voice bot: Transcribe (STT) + Bedrock + Polly (TTS) + Lambda.
AgentCore and managed-service rates last verified 14 July 2026 (us-east-1). Bedrock model prices last reviewed 28 July 2026. Estimates for planning, not a quote.