AI agents
Amazon Bedrock Agents Classic / Amazon Bedrock AgentCore vs Foundry Agent Service
Bedrock Agents Classic is in maintenance (no new customers since July 30, 2026).
On Azure, the equivalent concept is a Foundry prompt agent, and on AWS the successor is the AgentCore managed harness.
Side-by-side comparison
| Aspect | Amazon Web Services | Microsoft Azure |
|---|---|---|
| Service | Amazon Bedrock Agents Classic Maintenance Amazon Bedrock AgentCore | Foundry Agent Service |
| Lifecycle | The original Bedrock Agents (launched Nov 2023), renamed Agents Classic. Not open to new customers from July 30, 2026. AWS recommends migrating to AgentCore. Generally available | Voice-based agents and some monitoring/eval/telephony capabilities are in preview. |
| What it is | Configuration-based agents with action groups and knowledge bases. Modular agent platform. Harness (managed agent loop), Runtime, Memory, Gateway (MCP), Identity, Code Interpreter, Browser, Observability, Evaluations, Optimization, Policy, Registry, Payments. Works with any framework and any model. | Managed agent platform. Prompt agents (declarative), voice-based prompt agents (Voice Live), and hosted agents (your container or zip, any framework), plus the Responses API for "ephemeral" agents. Shared toolbox via a managed MCP endpoint. |
| Pricing model | Agents Classic charges no separate agent fee. You pay for model tokens and the resources invoked. | Prompt agents are billed per call for inference plus tool usage. |
| Docs | Amazon Bedrock Agents Classic docs Amazon Bedrock AgentCore docs | Foundry Agent Service docs |
| Pricing page | Amazon Bedrock Agents Classic pricing Amazon Bedrock AgentCore pricing | Foundry Agent Service pricing |
| Exams | Generative AI Developer – Professional | AI-103 |
Key differences
- AWS's migration path is the AgentCore harness. Action groups become MCP tools through AgentCore Gateway, and knowledge bases become gateway-fronted tools, with memory, guardrails, and tracing added.
- Foundry prompt agents are the declarative counterpart. They support instructions, a model, and tools (file search, code interpreter, web search, MCP, functions).
Pricing summary
Pricing models only. No prices are quoted because they change by region and over time.
- AWS
- Agents Classic charges no separate agent fee. You pay for model tokens and the resources invoked. Amazon Bedrock Agents Classic pricing → Amazon Bedrock AgentCore pricing →
- Azure
- Prompt agents are billed per call for inference plus tool usage. Foundry Agent Service pricing →
Certification exam mapping
| Exam | Provider | Level | Status |
|---|---|---|---|
| AWS Certified Generative AI Developer – Professional | AWS | Professional | current Exam code not confirmed from the official page at build time; shown by name only. |
| Microsoft Certified – Azure AI Apps and Agents Developer Associate AI-103 | Azure | Associate | current Generative AI and agentic solutions with Microsoft Foundry using Python. Successor in scope to the retired AI-102. |
Relevance is based on each exam's published scope, not a line-by-line objective match. Check the official study guide.
Verification notes
- "No separate agent fee" for Agents Classic reflects historical Bedrock pricing. Confirm on the pricing page.
FAQ
What is the Azure equivalent of Amazon Bedrock Agents Classic and Amazon Bedrock AgentCore?
Bedrock Agents Classic is in maintenance (no new customers since July 30, 2026). On Azure, the equivalent concept is a Foundry prompt agent, and on AWS the successor is the AgentCore managed harness.
How does pricing compare between Amazon Bedrock Agents Classic and Amazon Bedrock AgentCore and Foundry Agent Service?
AWS: Agents Classic charges no separate agent fee. You pay for model tokens and the resources invoked. Azure: Prompt agents are billed per call for inference plus tool usage. Check the official pricing pages for current rates.
Which certification exams cover this topic?
AWS Certified Generative AI Developer – Professional; Microsoft Certified – Azure AI Apps and Agents Developer Associate (AI-103).