GitLab ties agents to Vertex

GitLab said on April 14 that its Duo Agent Platform can now use Google Cloud Vertex AI models without moving agent actions outside GitLab’s own controls. (about.gitlab.com) The update lets GitLab agents call foundation models through Vertex AI, including Gemini models, while GitLab keeps compliance controls and audit trails around what those agents do inside software workflows. (about.gitlab.com) GitLab also said customers can run its artificial intelligence gateway on Google Cloud and use Vertex AI “with no separate AI infrastructure to provision or manage,” tying the setup to existing cloud access and policy controls. (about.gitlab.com) An agent platform is software that lets specialized bots plan, write, review, and fix code across a development pipeline. GitLab describes Duo Agent Platform as an orchestration layer for planning, coding, security review, and remediation across the software lifecycle. (about.gitlab.com) Vertex AI is Google Cloud’s managed service for building and using artificial intelligence models, and Google says it gives customers access to Gemini and more than 200 foundation models. (cloud.google.com) GitLab put Duo Agent Platform into general availability on January 15, 2026, for Premium and Ultimate customers on GitLab.com and self-managed deployments, after pitching the product as a way to spread artificial intelligence beyond code generation into testing, security, and delivery work. (about.gitlab.com) The company had already been pushing governed deployments for regulated customers. In February, GitLab said version 18.9 added self-hosted Duo Agent Platform support and Bring Your Own Model options for enterprises that want tighter control over where models run. (about.gitlab.com) Google Cloud has been making a similar enterprise pitch around control and oversight. Its Vertex AI documentation says the service uses Google Cloud security controls, while Google’s Agent Builder product page says customers can build and govern enterprise-grade agents on the platform. (cloud.google.com, cloud.google.com) The partnership extends a broader GitLab-Google Cloud relationship that predates this week’s announcement. Google Cloud wrote in 2024 that the two companies were already linking GitLab’s source code, continuous integration, and deployment tools with Google Cloud infrastructure. (cloud.google.com) For GitLab, the new pitch is narrower and more concrete: let teams use Google’s model layer through Vertex AI, but keep the agent’s work inside the same GitLab system that already tracks approvals, policies, and audit history. (about.gitlab.com)