GitLab Inc. announced an integration with Amazon Web Services that allows customers to route GitLab Duo Agent Platform inference through Amazon Bedrock models in their existing AWS accounts.

The integration enables joint customers to use Amazon Bedrock models, IAM policies, and AWS spending commitments already in place without requiring new vendor onboarding or separate model endpoints. GitLab Credits purchased through AWS Marketplace count toward existing AWS spending commitments.

The collaboration builds on GitLab's Bring Your Own Model capability for Self-Managed customers, which connects self-hosted AI Gateway directly to Amazon Bedrock. This keeps inference traffic and source code within customers' own AWS environments.

GitLab Duo Agent Platform uses a per-request billing model from a shared organizational pool of GitLab Credits, without per-seat pricing. For customers purchasing through AWS Marketplace, these credits apply to existing AWS spend commitments.

"GitLab Duo Agent Platform on Amazon Bedrock means customers can deploy agentic AI for software development without standing up new infrastructure, negotiating new contracts, or rethinking their security posture," said Rahul Pathak, vice president of data and AI GTM at AWS.

The platform adds workflow-level governance on top of Amazon Bedrock's model-level enforcement. GitLab serves as the system of record for merge requests, pipelines, and security findings, with admin-defined policies controlling which models agents can access and audit logs capturing agent actions.

Yoshiki Matsuda, chief operating officer at Fixstars Corporation, said the integration allows his organization to apply AI across their software delivery lifecycle while keeping inference traffic and code within their AWS environment.

The information is based on a GitLab press release statement.