GitLab Added Microsoft Foundry Support for AI

The update enables organizations to deploy specific AI models within their own Azure infrastructure for better data control.

Updated on Sept. 22, 2026 in Artificial Intelligence

Isometric editorial illustration of a sculptural server unit with an internal glowing data conduit, representing secure AI infrastructure deployment. — GitLab has integrated Microsoft Foundry support into its self-hosted platform, allowing organizations to deploy AI models within their own Azure infrastructure to enhance data sovereignty. AI Illustration. Upload story photo >

Live Poll

Should organizations prioritize controlling their own AI infrastructure over using pre-managed services?

GitLab has updated its GitLab Duo Self-Hosted platform to integrate with Microsoft Foundry. This development allows organizations to maintain their AI models within their own Azure environments.

Why it matters

The change addresses growing organizational demands for data sovereignty, regulatory compliance, and network isolation. It enables teams to route AI workloads through their own infrastructure rather than relying on external services.

The architecture comprises three distinct components: a self-managed GitLab instance, a self-hosted AI Gateway, and Microsoft Foundry endpoints. This setup allows users to map different OpenAI GPT, Anthropic Claude, Meta Llama, or Mistral models to specific GitLab Duo capabilities.

The players

GitLab

A software company providing a DevSecOps platform used for software development, security, and operations lifecycle management.

Microsoft

A multinational technology corporation offering the Azure cloud computing platform and the Microsoft Foundry model hosting service.

The details

The AI Gateway acts as an intermediary layer between GitLab Duo and selected model endpoints. By routing requests through this gateway, administrators keep AI traffic within their managed infrastructure. This configuration necessitates that teams oversee their own model deployment, compute capacity, and networking requirements to ensure performance.

September 22, 2026: GitLab announced the integration with Microsoft Foundry.

The Tech Race

This integration follows the industry-wide trend of bringing AI model inference closer to sensitive data stores. It positions GitLab to compete for enterprise customers who previously blocked AI adoption due to strict data residency requirements.

IT administrators and developers gain the ability to choose and manage specific AI models like Llama or Claude directly within their Azure environments. The shift requires teams to assume responsibility for infrastructure capacity and networking overhead.

The takeaway

Organizations looking to implement this should evaluate their internal Azure capacity and model requirements ahead of any deployment. Watch for future performance benchmarks related to the AI Gateway as more companies migrate from managed to self-hosted AI configurations.