Note
Access to this page requires authorization. You can try signing in or changing directories.
Access to this page requires authorization. You can try changing directories.
Azure Container Apps express provides the fastest way to deploy containerized web applications to Azure. With opinionated defaults and a minimal configuration surface, express is a developer-first and agent-first platform designed to get your web apps running in the cloud as fast as possible.
By using express, you can create a container app directly without waiting for environment provisioning first. The rapid provisioning and scale-from-zero features make express an ideal host for AI-powered applications and agent backends.
Key capabilities
The express deployment model removes infrastructure decisions so you can focus on building your application.
| Capability | Details |
|---|---|
| High-speed launch | Deploy in minutes with no infrastructure tuning required. Scaling behavior is built in from the start. |
| Simple, powerful apps | Run HTTP-first workloads including APIs, SaaS frontends, AI gateways, and event-driven web backends. |
| Automatic elasticity | Scale from zero to hyperscale automatically. The platform is designed for unpredictable traffic patterns, and scaling is handled for you. |
| Scale from zero | Your app scales down to zero when idle and back up on demand, so you only pay for what you use. |
| High-speed startup | Optimized cold start ensures your app is ready to serve traffic quickly. |
| Opinionated defaults | Sensible defaults are applied automatically so you don't have to configure infrastructure settings. |
| Minimal configuration surface | Fewer decisions to make means faster time to production. |
| Developer velocity | Spend less time on infrastructure and more time writing code. |
Common scenarios
The express deployment model works best for HTTP-based web workloads where speed of deployment and simplicity matter most.
SaaS applications: Launch SaaS products without worrying about scaling infrastructure.
AI app frontends: Deploy AI-powered interfaces and gateways that scale with demand.
Developer tools: Ship internal and external dev tools with zero-config deployment.
Web dashboards: Build internal analytics, monitoring, and admin panels with instant availability.
Startups and new projects: Go from idea to production in minutes. Prototype fast, and scale as you grow.
Rapid prototyping: Build and validate ideas quickly, then keep running in production without replatforming.
How express works
The express deployment model simplifies the deployment experience by removing the need to manage a Container Apps environment. You deploy your app and the platform provisions the underlying resources for you.
Less environment details to manage: When using the portal, the platform creates a light-weight environment for your app. If working from the CLI, you still create an environment yourself.
Consumption-based compute: Express apps run on consumption CPU with pay-as-you-go pricing. Your apps scale to zero when idle, so you only pay for the compute your app uses.
Opinionated defaults: The platform handles configuration decisions like scaling rules, networking, and resource allocation with production-ready defaults.
Request-driven duration: Compute runs when your app receives requests and scales down when traffic stops.
Optimized cold start: The platform automatically optimizes cold-start behavior, so your app is ready to serve traffic quickly after scaling from zero.
Specialized management UI: Express apps are managed through a UI experience separate from the Azure portal. When you create or manage an express app, the platform directs you to this streamlined interface instead of the standard Azure portal experience.
When to use express
Use the following table to determine if express is the right fit for your workload.
| Scenario | Use express | Alternative |
|---|---|---|
| Web apps and REST APIs | ✅ Yes | |
| SaaS frontends and AI gateways | ✅ Yes | |
| Rapid prototyping and startups | ✅ Yes | |
| Web dashboards and admin panels | ✅ Yes | |
| GPU workloads | ❌ No | Use serverless GPUs with dedicated workload profiles |
| TCP-based services | ❌ No | Use standard Container Apps environments |
| Jobs and batch processing | ❌ No | Use Container Apps jobs |
| Microservices with service discovery | ❌ No | Use standard Container Apps environments |
Considerations
Keep these important points in mind when using express:
HTTP workloads only: Express supports web apps and APIs that communicate over HTTP. TCP-based workloads aren't supported.
Consumption CPU compute: Express apps run on consumption-based CPU compute. GPU workloads require standard Container Apps with dedicated workload profiles.
Opinionated configuration: The express model uses opinionated defaults with a minimal configuration surface. If you need fine-grained control over compute, networking, or cold-start behavior, use standard Container Apps with a workload profiles environment.
Feature availability: Express offers a focused set of features at launch. Some capabilities available in standard Container Apps environments, such as custom virtual networks, Dapr integration, and built-in service discovery, aren't available in express.
Supported features
The express deployment model supports the following features. This list is updated as Microsoft enables new capabilities.
| Feature | Supported |
|---|---|
| Scale to zero | ✅ Yes |
| Image deployment (anonymous and token-based) | ✅ Yes |
| Multi replica | ✅ Yes |
| Environment variables | ✅ Yes |
| Enable ingress | ✅ Yes |
| New portal experience | ✅ Yes |
| Log streaming | ✅ Yes |
| Region restriction | ✅ Yes |
| Logs (Log Analytics) | ✅ Yes |
| Rolling updates | ⚠️ Partial |
| Secrets | ❌ No |
| Billing | ❌ No |
| Secrets from Key Vault | ❌ No |
| Autoscale (KEDA-based) | ❌ No |
| Managed identity (app runtime) | ❌ No |
| Managed identity (image pull) | ❌ No |
| VNet integration | ❌ No |
| Quota | ❌ No |
| Health probes | ❌ No |
| Exec access | ❌ No |
| Easy Auth | ❌ No |
| Metrics (Azure Monitor) | ❌ No |
| Custom domain (managed certificate) | ❌ No |
| IP restrictions | ❌ No |
| CORS | ❌ No |
| Logs (Azure Monitor) | ❌ No |
| Session affinity | ❌ No |
| Sidecar container | ❌ No |
| Init container | ❌ No |
| Volume mount | ❌ No |
| Ephemeral storage | ❌ No |
| GPU | ❌ No |
| Insecure HTTP ingress | ❌ No |
| Additional ports | ❌ No |
| App-to-app communication | ❌ No |
| Debug console | ❌ No |
| Deployment label | ❌ No |
| Language stack | ❌ No |
| Multi revision / traffic splitting | ❌ No |
| Resiliency | ❌ No |
| Source-to-cloud deployment | ❌ No |
| TCP protocol | ❌ No |
| Aspire | ❌ No |
| Maintenance window | ❌ No |
| OpenTelemetry | ❌ No |
| Premium ingress | ❌ No |
| Private endpoint | ❌ No |
| Workload profiles | ❌ No |
| Peer-to-peer encryption | ❌ No |
| Job | ❌ No |
| Single revision management | ❌ No |
| Custom domain (BYOC) | ❌ No |
| Environment custom domain suffix (BYOC) | ❌ No |
| Environment custom domain suffix (managed certificate) | ❌ No |
| Azure file storage | ❌ No |
| Zone redundancy | ❌ No |
| App-to-app (internal FQDN) | ❌ No |
| Internal vs. external apps | ❌ No |
Region availability
During public preview, express is initially available in the West Central US and East Asia regions. Support for more regions will be coming soon.