GitLab Blog
AI news source. Articles are auto-selected and adapted by Hamidun News editors.
Latest publications
Claude Sonnet 5 Became First Model to Pass All GitLab Duo Agent Platform Tests
Anthropic released Claude Sonnet 5 on GitLab Duo Agent Platform — the first model to pass all GitLab benchmark tasks, versus 93.8% for Sonnet 4.6.
GitLab Migrated Rate Limiting to AI Agents: 14 Specs and 30+ Merge Requests
A small team at GitLab used AI agents and GitLab Duo to migrate a legacy rate limiting system, producing 14 specifications and over 30 merge requests.
GitLab Improved Restricted Access Feature for License Spending Control
GitLab improved the Restricted Access feature: it blocks adding new paid users when licenses are exhausted and now works better with automatic provisioning via SAML, SCIM, and LDAP.
Claude Fable 5 arrives on GitLab Duo — multi-day agent tasks without manual stops
Anthropic brought a Mythos-class model to GitLab Duo Agent Platform — Fable 5 completes multi-stage tasks on its own, without manual checkpoints or repeated instructions.
GitLab launched a managed DevSecOps platform on Google Cloud with Gemini 3.5 Flash
GitLab now runs as a fully managed platform on Google Cloud, with Gemini 3.5 Flash and Gemma 4 built in and full control over the company’s code and data.
GitLab 19.1: AI workflows now trigger automatically on events, with no human involvement
GitLab 19.1 adds event-based triggers for Duo Flows, administrative control over agents, and configuration validation — enabling continuous, governed AI workflow execution for enterprise.
GitLab Orbit lands in Google Antigravity: AI agents gain access to full-project context
GitLab Orbit is now available in Google Antigravity's MCP Store, giving AI agents context across the entire development lifecycle without switching between tools.
Claude Opus 4.8 in GitLab: Precision in Complex Multi-Step Tasks
Anthropic released Opus 4.8 for autonomous agents on complex projects in GitLab Duo. Now supports instruction updates mid-session without cache reloads.
GitLab revealed why AI coders need platform context, not just code
Every week new demos of AI agents emerge that turn a prompt into a pull request in five minutes. But demos end after the commit. The PR doesn't reference the issue it was supposed to fix.
GitLab Expanded Support for Open-Source Models for Closed Networks
GitLab 19.0 added four new open-source models for teams with strict data residency requirements. Models include Mistral Devstral, GLM-5.1, Kimi-K2.6, and MiniMax-M2.7 for on-premises deployment.
GitLab Automates Full Merge Request Lifecycle with Developer Flow
GitLab introduced Developer Flow — an AI agent that now manages the entire merge request process: from analyzing reviewer feedback to resolving conflicts and final merge.
Codex and GitLab: From Code to Production in Three Steps
GitLab demonstrated how the AI agent Codex can write code in the terminal, but delivering it to production is a job for the entire platform. Three scenarios from local bug fixes to merge requests.
GitLab Duo CLI: managing AI agents in CI/CD pipelines
When an AI agent runs in CI/CD without human approval, built-in control systems are needed. GitLab addresses this at the platform level with governance controls.