What's new in GitLab

Monthly releases, developer resources, and the latest from the GitLab team.

Monthly releases

April 16, 2026

GitLab 18.11: Extend agentic AI across the SDLC with confidence and control

Autonomous remediation driven by the need to reduce risk in software development, two new foundational agents for CI and analytics, budget guardrails and usage caps for GitLab Credits, and more.

Read the release notes

Blog

Automate remediation with ready-to-merge AI code fixes

With GitLab 18.11, Agentic SAST Vulnerability Resolution becomes generally available, alleviating security bottlenecks.

Blog

CI Expert and Data Analyst AI agents target development gaps

Set up CI and query your software development lifecycle data with two new GitLab Duo Agent Platform foundational agents available in GitLab 18.11.

Blog

Budget guardrails for GitLab Credits

Learn how new spending caps and per-user credit limits give organizations the budget guardrails to scale GitLab Duo Agent Platform.

Blog

Claude Opus 4.7 is now available in GitLab Duo Agent Platform

Anthropic's latest model, available now, for stronger agent work.

This month's Notable Contributor: Rinku C

We are excited to recognize Rinku C, a Level 4 contributor with over 80 merged improvements across GitLab since joining in September 2025.

Read the press release

Secure

Agentic SAST Vulnerability Resolution (GA)

Agentic vulnerability resolution uses iterative reasoning to analyze vulnerability context across the codebase, generate and validate high-quality fixes, and provide confidence scoring for proposed solutions.

Manage

Claude Opus 4.7 now available in GitLab Duo Agent Platform

Anthropic's latest model is now available in GitLab Duo Agent Platform via model selection in Agentic Chat and across agent-powered workflows.

Analyze

Data Analyst Agent (GA)

The Data Analyst Agent covers merge requests, issues, projects, pipelines, and jobs, querying data already in GitLab via natural language, without requiring GLQL knowledge or a dashboard request.

Manage

Budget and usage guardrails for GitLab Credits

Organizations scaling their use of agentic AI across the software lifecycle need cost predictability as they expand adoption. Spending caps for GitLab Credits put technology leaders in direct control.

Build

CI Expert Agent (Beta)

The CI Expert Agent inspects your repository, detects your language and framework, and proposes a working build and test pipeline with plain-language explanations for every decision.

Manage

Mistral AI added as self-hosted model in GitLab Duo Agent Platform

GitLab Duo Agent Platform now supports Mistral AI as an LLM platform for self-hosted model deployments. GitLab Self-Managed customers can configure Mistral AI models through the AI Gateway for use with agents and flows.

Secure

Risk-based security prioritization and enforcement

When severity scores don’t reflect real exploitability, developers stop trusting the signal and start ignoring it. This capability reduces triage work and ensures that developers focus on the findings that matter.

Manage

Fine-grained permissions for personal access tokens (Beta)

Limit each personal access token (PAT) to specific resources and actions, reducing the potential damage of a leaked or compromised token. Fine-grained PATs are available across all tiers, including Free.

Manage

Service accounts: now on Free and available in subgroups and projects

Service accounts are now available on GitLab.com across all tiers, including Free (up to 100 per top-level group). Teams can also create service accounts directly in subgroups and projects with scoped access.

Analyze

Manage

ClickHouse integration generally available for self-managed deployments

Self-managed instances can now use ClickHouse as a production-ready analytics backend, powering dashboards and API endpoints that require high-performance data querying at scale.

Code

GitLab now available on the Cursor marketplace

GitLab is now available as a plugin on the Cursor marketplace, connecting Cursor to your GitLab instance via the GitLab MCP server.

Code

Build

Reconfigure inputs when manually running MR pipelines

CI/CD inputs let you customize pipeline behavior at runtime, now that flexibility extends to merge request pipelines, providing the same runtime control across all pipeline types.

Read the press release