Originally published at ssojet
Docker Desktop 4.40 Features
Docker Desktop 4.40, released on March 31, 2025, introduces significant enhancements for AI development workflows, including the new Docker Model Runner. This feature allows developers to pull, run, and manage AI models directly from Docker Hub within Docker Desktop, streamlining the integration of AI models into containerized applications. It simplifies the traditionally complex process of model setup and testing.
Image courtesy of Docker
Developers can now execute models locally, benefiting from GPU acceleration on Apple Silicon for improved performance. Standard model packaging using OCI Artifacts allows models to be versioned, stored, and shipped using familiar CI/CD workflows. For further details, check the Docker Model Runner documentation.
AI Agent Enhancements
The Docker AI Agent has also received notable upgrades, including support for the Model Context Protocol (MCP). This integration enables seamless interaction with external tools and data sources, allowing developers to manage local files, run shell commands, and conduct Git operations directly through the AI Agent. The AI Tool Catalog extension simplifies the process of discovering and connecting to various MCP servers, enhancing the modularity of the development environment.
Image courtesy of Docker
For enterprise users, Docker Desktop 4.40 introduces Settings Reporting, which provides administrators with comprehensive visibility into user compliance with assigned settings policies. This feature allows quick identification of non-compliance issues, enhancing organizational security and compliance efforts.
Docker MCP Integration
Docker has embraced the Model Context Protocol (MCP) to provide a unified approach for AI development. The Docker MCP Catalog allows developers to discover and manage over 100 MCP servers integrated within Docker Hub. This centralized approach simplifies the deployment of AI applications by making it easier to retrieve data from external sources, interact with local file systems, and collaborate with third-party services.
Image courtesy of Docker
With tools like Docker Scout for security analysis and the Docker MCP Toolkit for enterprise-ready development, Docker aims to enhance the AI developer experience. The MCP protocol, originally developed by Anthropic, is designed to streamline AI application development without forcing developers to abandon their existing tools.
Local AI Model Development
The Docker Model Runner enables local execution of AI models, making it as simple as running a container. This is particularly beneficial for developers using Apple Mac laptops, allowing GPU acceleration for faster performance. Models are packaged as OCI Artifacts, facilitating distribution through Docker Hub or internal registries.
Image courtesy of SiliconANGLE
Docker is also collaborating with companies like Google, Qualcomm, and HuggingFace to provide a comprehensive ecosystem of AI tools within the Model Runner. This partnership aims to enhance developer productivity by integrating familiar AI frameworks and tools directly into the Docker environment.
Enterprise Compliance and User Management
Docker Desktop 4.40's Settings Reporting feature is crucial for enterprise clients, offering tools to monitor compliance with assigned settings policies. This capability is essential for organizations looking to enforce security and compliance standards across development teams. Docker’s commitment to enterprise-grade management tools aligns well with the needs of organizations prioritizing secure software delivery.
For businesses, implementing secure Single Sign-On (SSO) and user management is critical. SSOJet offers an API-first platform featuring directory sync, SAML, OIDC, and magic link authentication, ensuring that enterprises can manage user identities efficiently and securely.
For further information about SSOJet’s services, visit SSOJet. Explore how SSOJet can enhance your authentication processes and support your development teams in creating secure applications.