GitLab Duo

Features

• Use specialized agents for planning, coding, security analysis, and analytics
• Create custom agents that follow organizational standards and workflows
• Run multi-step agentic flows triggered by GitLab events
• Use agentic chat with project and software-lifecycle context
• Connect internal systems, third-party tools, and external AI services
• Apply policy controls, permissions, model selection, and session traceability
• Use self-hosted language models with GitLab Self-Managed

Use cases

• Plan and prioritize software development work
• Review code and suggest or implement fixes
• Diagnose and repair failed CI/CD pipelines
• Automate security analysis and vulnerability workflows
• Create repeatable workflows triggered by issues, mentions, or pipelines
• Connect AI agents to internal tools and software data

Pros

Cons

Latest updates

• 2026-09-17 GitLab 19.4 release notes (19.4)

  GitLab Duo Agent Platform tools can be governed with rules for GitLab MCP server tools.

• 2026-08-20 GitLab 19.3 release notes (19.3)

  The Flow Creator foundational agent is available in the AI Catalog to create custom flows.

• 2026-07-16 GitLab 19.2 release notes (19.2)

  GitLab Duo CLI is generally available.

• 2026-06-18 GitLab 19.1 release notes (19.1)

  Secret false positive detection with the GitLab Duo Agent Platform is generally available.

• 2026-05-21 GitLab 19.0 release notes (19.0)

  Custom review instructions for GitLab Duo can be defined at the group level.

Capabilities

• Inline completion — “AI Code Suggestions in the IDE (Duo Classic)” source
• Chat about your code — “AI Chat in the IDE (Duo Classic)” source
• Agent that edits files — “Automatically transform issues into production-ready code.” source
• Knows the whole codebase — “every agent can reason across your full software lifecycle rather than just the files in its immediate context.” source
• Reviews pull requests — “Gather feedback, iterate on merge requests, and more.” source
• Choice of models — “Choose the right LLM for your agents, flows and agentic chat to power your use cases.” source
• Self-hosted — “Self-Managed runs in your environment on any cloud or on premises, including air-gapped patterns.” source
• API — “Use APIs, webhooks, integrations, and Model Context Protocol (MCP) so agents and tools share context under your policies.” source
• Builds agents and workflows — “Create, test, and share custom agents with specific skills.” source
• Traces and evaluates — “Flow sessions are visible in GitLab and include trigger details and links to CI/CD job logs, giving teams clear insight into how automation is executed.” source

Get it

• Documentation

Security

• SOC 2 Type II — “GitLab maintains a SOC 2 Type 2 report for the Security, Confidentiality and Availability Trust Services Criteria for GitLab.com.” source
• SOC 2 — “GitLab maintains a SOC 2 Type 2 report for the Security, Confidentiality and Availability Trust Services Criteria for GitLab.com.” source
• ISO 27001 — “GitLab maintains ISO/IEC 27001:2022 certification for the information security management system supporting the GitLab software-as-a-service (SaaS) subscriptions, GitLab.com and GitLab Dedicated.” source
• GDPR — “GitLab is compliant with GDPR requirements.” source

Pricing

Starting price

$29/user/mo

Prices checked

2026-09-25

Free

Free

• 5 users per top-level group
• 400 compute minutes per month
• Source Code Management & CI/CD
• 10 GiB adjustable storage
• Static Application Security Testing

Premium

• $29 per user/month, billed annually

• Unlimited licensed users
• 10,000 compute minutes per month
• Advanced CI/CD
• Team Project Management
• SLA Management
• Priority Support

Ultimate

Price on request

• Unlimited guest users
• 50,000 compute minutes per month
• Application Security Testing
• Software Supply Chain Security
• Vulnerability Management
• Strategic Portfolio Management