GitLab ties agents to Google Cloud

GitLab has tied its Duo Agent Platform to Google Cloud’s Vertex AI, letting software teams run GitLab agents on Google’s models inside GitLab’s own control system. (about.gitlab.com) GitLab said on April 14 that Duo agents can now call foundation models through Vertex AI, including Gemini, while agent actions stay subject to GitLab access controls, approval policies, compliance rules, and audit logs. (about.gitlab.com) The practical change is that customers can run GitLab’s artificial intelligence gateway on Google Cloud without setting up separate artificial intelligence infrastructure, and GitLab said usage can count toward existing Google Cloud commitments. (about.gitlab.com; financialcontent.com) An artificial intelligence agent is software that can take on a task, such as reviewing code or investigating a security issue, with limited human prompting. GitLab says its Duo Agent Platform places those agents across the software development lifecycle so teams can delegate coding, refactoring, security scans, and research. (docs.gitlab.com; docs.gitlab.com) Vertex AI is Google Cloud’s managed platform for building and using generative artificial intelligence systems, and Google says it gives customers access to Gemini and more than 200 foundation models. That makes GitLab’s new link less about adding a chatbot and more about plugging a software workflow into a catalog of models already sold through Google Cloud. (cloud.google.com; about.gitlab.com) GitLab launched the Duo Agent Platform into general availability on January 15, 2026, for Premium and Ultimate customers on GitLab.com and self-managed deployments. In its Google Cloud blog post this week, GitLab said the platform acts as an orchestration layer where teams and specialized agents plan, code, review, and remediate security vulnerabilities together. (about.gitlab.com; about.gitlab.com) GitLab’s documentation says the artificial intelligence gateway is the service layer that routes requests to model providers, including Google Vertex AI. The same documentation says those third-party providers have their own data-handling practices, which is one reason large companies focus on where models run and how requests are governed. (docs.gitlab.com) GitLab framed the Google tie-up as a way for enterprises to use models they already buy while keeping software delivery work inside one governed system. Google has been pushing Vertex AI and Agent Builder as enterprise tools for building and governing agents, so the partnership also gives Google another route into corporate engineering teams. (about.gitlab.com; cloud.google.com) The pitch is straightforward: keep the agents in GitLab, keep the models in Google Cloud, and keep the approvals, logs, and spending inside systems enterprise buyers already use. (about.gitlab.com; about.gitlab.com)