Claude Sonnet 5 on GitLab: More reliable, more efficient

2026-06-30 · Source: GitLab · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering · Depth: Intermediate, quick

Summary

Anthropic's Claude Sonnet 5 is now integrated into the GitLab Duo Agent Platform across all tiers and deployment models via GitLab's AI Gateway. This new model is designed for multi-step tasks, generating high-quality code, and conducting workflows affordably at scale. It is the first model in GitLab's evaluation suite to complete all benchmark tasks, a significant improvement over its predecessor, Sonnet 4.6, which completed 93.8%. This enhanced reliability translates to 8.8% more issues resolved and more usable output for teams utilizing GitLab Duo agents, improving daily workflows like multi-file refactoring, test generation, and security investigations. The model also offers improved efficiency, lowering the real cost of completed tasks by finishing more work with fewer resources, complementing other models like Claude Opus 4.8 for complex tasks.

Key takeaway

For AI Engineers and ML Directors managing software development workflows on GitLab Duo, the integration of Claude Sonnet 5 means you can delegate more multi-step coding tasks with confidence. Its 100% benchmark completion and 8.8% higher issue resolution rate directly improve output quality and reduce intervention needs. Evaluate Sonnet 5 as your default for everyday development work to enhance efficiency and affordability, reserving Claude Opus 4.8 for the most complex reasoning tasks.

Key insights

Claude Sonnet 5 significantly enhances agent reliability and efficiency on GitLab Duo for software development tasks.

Principles

• Agent reliability enables delegation, not just supervision.
• Efficiency and reliability compound task cost savings.
• Model selection should match task complexity and cost.

Method

The article describes integrating Claude Sonnet 5 into GitLab Duo Agent Platform via AI Gateway, allowing model selection per task for varied development workflows.

In practice

• Use Sonnet 5 for multi-file refactoring.
• Generate test coverage with Sonnet 5.
• Trace security investigations across repositories.

Topics

• Claude Sonnet 5
• GitLab Duo Agent Platform
• AI Gateway
• Code Generation
• Agentic AI
• Software Development Workflows

Best for: CTO, VP of Engineering/Data, AI Architect, AI Engineer, Machine Learning Engineer, Director of AI/ML

Related on AIssential

• Claude Sonnet 5: The Fable 5 at Home — Claude Sonnet 5 offers enhanced agentic capabilities, self-verification, and tool use, making it a cost-effective default for complex daily tasks.
• Claude Opus 4.8 on GitLab: Complex agentic work, less disruption — Claude Opus 4.8 on GitLab Duo Agent Platform streamlines complex multi-step agent workflows by minimizing disruptions.
• Claude Sonnet 5: Greatest AI Coding Model Ever! 1M Context, Cheap, & More! (Early Test) — Claude Sonnet 5 offers best-in-class agentic coding with massive context and competitive pricing.
• Claude Sonnet 5 in 12 mins! — Claude Sonnet 5 provides strong code generation capabilities but at a higher token and monetary cost than GLM 5.2.
• Introducing Claude Sonnet 5 on AWS: Anthropic’s most capable Sonnet model — Claude Sonnet 5 delivers near-Opus intelligence at Sonnet pricing, enhancing reliability for coding, agents, and professional tasks.
• A Product Take on Sonnet 4.5 — Sonnet 4.5 significantly improves upon Opus 4.1's coding capabilities, demonstrating higher "wisdom" and efficiency.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by GitLab.