Build AI.On solid ground.
Reference architectures and deployable Azure foundations for your AI applications and agents.
Bicep and Terraform · Built on Azure Verified Modules
Choose your landing zone.
Agent Landing Zone
An Azure foundation for AI applications and agents, with a default application built on Microsoft Foundry.
AI Gateway Landing Zone
Shared model access for AI applications and agents, using Azure API Management.
How the landing zones connect
Reference architectures.
Explore Agent Landing Zone, integrated with AI Gateway Landing Zone, or on its own.
Agent Landing Zone
Agent Landing Zone connects to a platform that provides shared connectivity, DNS and egress services.
Enlarge the Agent Landing Zone diagram
Resource selection and reuse depend on your implementation and settings. This reference shows the default hosted application without application Cosmos DB or its private endpoint; Foundry Agent Service's separate backing resources are retained.
Agent Landing Zone with AI Gateway
Combine many Agent Landing Zones as spokes for agents to an AI Gateway Landing Zone as hub for models. The gateway has its own deployment and connectivity requirements; Agent Landing Zone does not provision it automatically.
Enlarge the integrated gateway diagram
Resource selection and reuse depend on your implementation and settings. The Agent Landing Zone portion shows the default hosted application without application Cosmos DB or its private endpoint; Foundry Agent Service's separate backing resources are retained.
Standalone Agent Landing Zone
Deploy an independent environment where your team manages the network, without a shared platform hub.
Enlarge the standalone diagram
Resource selection and reuse depend on your implementation and settings. This reference shows the default hosted application without application Cosmos DB or its private endpoint; Foundry Agent Service's separate backing resources are retained.
Choose your implementation.
| Implementation | Bicep | Terraform |
|---|---|---|
| Agent Landing Zone | ||
| AI Gateway Landing Zone | View implementation | View implementation |
| Foundry AVM modules | Explore the module | Explore the module |
Prefer Azure Portal? Open the Agent Landing Zone deployment.
Designed and tested for Azure public cloud. Validate government and sovereign deployments separately. Some included services may be in preview.
Additional implementation references
The Bicep implementation and Agent Landing Zone Terraform reference provide additional implementation details. The shared AI Landing Zone Terraform foundation reference is separate from Agent Landing Zone's integrated application lifecycle.
What will you build?
Knowledge assistants
Help employees find answers in organizational policies, procedures and technical documentation.
Document processing
Extract, classify and summarize information from invoices, contracts and reports.
Business process agents
Coordinate tasks across business systems, such as service requests and approval workflows.
Design guidance.
Checklist and architecture guidance for your Azure environment.
Supporting diagrams
Design areas
Cloud Adoption Framework
Well-Architected Framework
More Azure guidance
Start with the CAF AI Scenario, AI adoption steps, and AI strategy. For generative and non-generative workloads, consult AI on Azure Platform and the resource selection guidance. For infrastructure-based AI workloads, see AI on Azure Infrastructure.
AI Landing Zones provides workload landing zone reference architectures and deployable implementations.
Project and legal information
Roadmap
AI Landing Zones continues to evolve with new Azure capabilities and workload requirements. Explore the project roadmap to follow planned improvements and upcoming work across the reference architectures and implementations.
Contributing
This project welcomes contributions and suggestions. Most contributions require you to agree to a Contributor License Agreement (CLA) declaring that you have the right to, and actually do, grant us the rights to use your contribution. For details, visit https://cla.opensource.microsoft.com.
When you submit a pull request, a CLA bot will automatically determine whether you need to provide a CLA and decorate the PR appropriately (e.g., status check, comment). Simply follow the instructions provided by the bot. You will only need to do this once across all repos using our CLA.
This project has adopted the Microsoft Open Source Code of Conduct. For more information see the Code of Conduct FAQ or contact opencode@microsoft.com with any additional questions or comments.
Trademarks
This project may contain trademarks or logos for projects, products, or services. Authorized use of Microsoft trademarks or logos is subject to and must follow Microsoft's Trademark & Brand Guidelines. Use of Microsoft trademarks or logos in modified versions of this project must not cause confusion or imply Microsoft sponsorship. Any use of third-party trademarks or logos are subject to those third-party's policies.