Home
News
Tech Grid
Interviews
Anecdotes
Think Stack
Press Releases
Articles
  • Agentic AI

GitLab 18.10: Affordable Agentic AI for All Teams


GitLab 18.10: Affordable Agentic AI for All Teams
  • by: Business Wire
  • |
  • March 20, 2026

GitLab has released version 18.10, significantly broadening and simplifying access to agentic AI capabilities through the GitLab Duo Agent Platform. This update tackles the shift in software development bottlenecks from code creation—now accelerated by AI—to the numerous subsequent engineering tasks. By lowering barriers to adoption, reducing costs, and building greater trust, GitLab 18.10 enables teams to apply intelligent automation across the entire software development lifecycle more effectively.

Quick Intel

  • GitLab 18.10 makes GitLab Duo Agent Platform available to free-tier users on GitLab.com via monthly GitLab Credits commitment
  • Credits are shared group-wide, providing per-member access to agentic AI without per-seat fees
  • Agentic Code Review priced at flat $0.25 per review (4 reviews per Credit), enabling scalable, predictable automated reviews
  • Agentic SAST false positive detection now generally available for Ultimate customers, scoring findings to cut alert fatigue
  • GitLab Credits dashboard offers visibility into AI usage and cost monitoring tied to software delivery
  • Enhances security, speed, and innovation while maintaining budget control and platform guardrails

With AI speeding up coding, the real challenges now lie in review, security scanning, deployment, and related processes that slow delivery. GitLab 18.10 introduces affordable, accessible agentic AI tools to remove these obstacles and support faster, more secure software shipping.

Broader Access to GitLab Duo Agent Platform on Free Tier

Organizations using the GitLab.com free tier can now activate the GitLab Duo Agent Platform by committing to a monthly purchase of GitLab Credits. These credits are pooled at the group level, granting immediate agentic AI access to all team members without requiring individual seat licenses or plan upgrades. The dedicated GitLab Credits dashboard provides group owners with transparent tracking of credit consumption by agents and workflows, linking AI activity directly to development outcomes and supporting effective budget oversight.

Affordable and Scalable Agentic Code Reviews

Manual code reviews often create significant delays, extending delivery cycles and hindering innovation. Agentic Code Review in GitLab 18.10 automates merge request evaluations with complete context from the repository, CI/CD pipelines, and security policies. Priced at a consistent $0.25 per review (currently 4 reviews per GitLab Credit), this feature delivers predictable costs regardless of volume. As merge request numbers grow, organizations can realize major time and expense reductions compared to traditional manual review efforts.

Generally Available Agentic False Positive Detection for Security Scanning

GitLab Duo Agent Platform now offers generally available SAST false positive detection exclusively for GitLab Ultimate customers. Following each SAST scan, the platform automatically assesses new critical and high-severity findings, assigns a false positive likelihood score, and displays this evaluation in the Vulnerability Report. This capability helps security and development teams prioritize actionable vulnerabilities, minimize alert fatigue, speed up remediation, and retain ultimate control over dismissal decisions.

GitLab 18.10 democratizes advanced agentic AI, allowing software teams of every size to increase innovation velocity securely, cost-effectively, and within their defined guardrails.

About GitLab

GitLab is the intelligent orchestration platform for DevSecOps. GitLab enables organizations to increase developer productivity, improve operational efficiency, reduce security and compliance risk, and accelerate digital transformation. More than 50 million registered users and 50% of the Fortune 100* trust GitLab to ship better, more secure software faster.

  • Agentic AIDev Sec OpsAI Code ReviewSoftware Development
News Disclaimer
  • Share