BERLIN, July 10, 2025 (GLOBE NEWSWIRE) -- Docker, Inc.®, a provider of cloud-native and AI-native development tools, infrastructure, and services, today announced major new capabilities that make it dramatically easier for developers to build, run, and scale intelligent, agentic applications. By extending Docker Compose to support agents and AI models, introducing Docker Offload for cloud-scale execution, and collaborating with cloud providers like Google Cloud and Microsoft Azure and AI SDKs like CrewAI, Embabel, LangGraph, Sema4.ai, Spring AI, and Vercel AI SDK, Docker is delivering on its mission to simplify complex technology and empower developers.
“Agentic applications are rapidly evolving, but building production-grade agentic systems is still too hard,” said Tushar Jain, EVP of Engineering, Docker, Inc. “Just like Docker democratized microservices a decade ago, we’re now making agentic apps accessible to every developer by making agent-based development as easy, secure, and repeatable as container-based app development has always been. The next wave of software is powered by intelligent agents, and Docker makes it easy to turn that potential into real, running applications.”
These advancements are more than just new tools. They help solve one of the biggest challenges facing developers today, which is moving agentic applications from local prototypes to secure and scalable production environments.
Docker Compose Enters the Agent Era
For over a decade, Docker Compose has been the go-to tool used by millions of developers for defining and running multi-container applications. Now, Docker is extending Compose into the agent era, enabling developers to define intelligent agent architectures consisting of models and tools in the same simple YAML files they already use for microservices and take those agents to production.
With the new Compose capabilities, developers can:
- Define agents, models, and tools as services in a single Compose file
- Run agentic workloads locally or deploy seamlessly to cloud services like Google Cloud Run or Azure Container Apps
- Integrate with Docker’s open source Model Context Protocol (MCP) Gateway for secure tool discovery and communication
- Share, version, and deploy agentic stacks across environments without rewriting infrastructure code
This approach brings powerful agent orchestration into familiar workflows with no new languages or tools required.
“Making it just as straightforward for developers to take AI apps from prototype into production as it already is for regular code—that’s the next big thing in app development,” said Torsten Volk, Principal Analyst at Enterprise Strategy Group. “Expanding Docker Compose to give developers the same familiar, simple experience for AI deployments as they have for traditional apps is exactly what we need. Plus, the new capability to run AI models directly in the cloud—without clogging up your laptop—is another major step forward. This should make a real difference in how quickly enterprises can start adopting AI at scale.”
Introducing Docker Offload: Cloud Power, Local Simplicity
As agentic applications demand more GPU power for complex AI tasks, local machines frequently fall short of the necessary capacity, which has become a significant pain point for developers. To solve this, Docker today unveiled Docker Offload (Beta), a new capability that enables developers to offload AI and GPU-intensive workloads to the cloud without disrupting their existing workflows.
With Docker Offload, developers can:
- Maintain local development speed while accessing cloud-scale compute and GPUs
- Run large models and multi-agent systems in high-performance cloud environments
- Choose where and when to offload workloads for privacy, cost, and performance optimization
- Keep data and workloads within specific regions to meet sovereignty requirements and ensure data does not leave designated zones across the globe.
Docker Offload integrates directly into Docker Desktop, preserving the familiar docker compose up experience while delivering cloud horsepower under the hood.
Built on a Thriving Ecosystem
Docker’s agentic capabilities are launching alongside new integrations with leading cloud and AI platforms. Key partnerships include:
- Google Cloud: Deploy agentic applications to production via serverless environments with the new gcloud compose up command
- Microsoft Azure: Seamless deployments via Azure Container Apps, arriving soon
- Popular agent frameworks: Compose integrations now support CrewAI, Embabel, Google’s ADK, LangGraph, Spring AI, and Vercel AI SDK, and more.
Steren Giannini, Director of Product Management, Google Cloud Run
“With Compose Spec support in Cloud Run, we’re making it dramatically simpler to move sophisticated AI apps from local development straight to production. This collaboration brings the best of both worlds: Docker's local dev power combined with Cloud Run's serverless scale and reliability, all with one simple command.”
Scott Hunter, Vice President Director of Product, Azure Developer Experience
“Microsoft has been collaborating with Docker for many years, and we’re pleased to see Docker extend the Compose Spec to support agent-based application development.
We’re working together to make agentic app deployment seamless on Microsoft Azure Container Apps—helping developers easily build and scale AI applications and agents from local dev to the cloud. These integrations help ensure developers can easily adopt Docker’s agentic tooling alongside the frameworks they already use.”
Ram Venkatesh, CTO and Co-founder of Sema4.ai
“Agent-based systems represent a transformative leap in how software interacts with the world: autonomous, goal-driven, and contextual. Docker’s direction supporting agentic architectures is a major unlock for developers, making it radically easier to compose, scale, and iterate on multi-agent systems without compromising security or reinventing infrastructure. This is the kind of pragmatic innovation that accelerates agentic adoption for real-world use cases.”
Craig McLuckie, CEO and Founder of Stacklok and Co-Creator of Kubernetes “Enterprises that want to lift up their knowledge workers and create powerful new customer experiences need to connect the right data to AI models at the right times, and that requires use of MCP. Docker has a critical role to play in facilitating adoption of MCP through familiar constructs like containers and Docker Compose. By working together, in the open, we can bring simplicity and security to MCP that will unlock real enterprise adoption.”
Availability
- Docker Compose enhancements for agentic applications are available today
- Docker Offload is available in closed beta for developers who request access.
- Google Cloud Run integration is live; Azure Container Apps support is coming soon
- MCP Gateway and Docker Hub MCP Server are open source and ready for use
Resources
- Blog: Docker Makes Building AI Agents Easy: Compose for Agent Era
- Video: Compose the Future: Build Agentic AI Apps with Docker
- Live Stream: Watch the keynote presentation at WeAreDevelopers at www.docker.com/WAD
- Google Cloud: From Localhost to Launch: Simplify AI App Deployment with Cloud Run and Compose
- Docker for AI Solutions https://www.docker.com/solutions/docker-ai/
- Docker Offload https://www.docker.com/products/docker-offload
- Docker Model Runner https://www.docker.com/products/model-runner/
- Demos GitHub Repos https://github.com/docker/compose-agents-demo
About Docker
Docker drives modern software development by making it easy to adopt container technology to radically boost productivity, security, testing, and collaboration at every step of the developer experience, including emerging AI workflows. Embraced by over 20 million developers worldwide, Docker’s unmatched flexibility and choice make it the preferred tool for developers seeking efficiency and innovation for creating modern applications. Learn more about Docker at www.docker.com.
A video accompanying this announcement is available at:
https://www.globenewswire.com/NewsRoom/AttachmentNg/972c3a61-d15d-40cf-a719-ad54e0a73fc2

Contact David Oro press@docker.com