Claude Sonnet 5 finishes multi-step agent workflows without stalling; GitLab reports 8.8% more issues resolved than Sonnet 4.6, reducing cost per completed task.
Summary
Agent task failures mid-execution waste time on diagnosis and re-prompting. A model that finishes reliably shifts developer effort from restarting runs to code review, making agents delegatable rather than supervisory.
Why it matters
Agent task failures mid-execution waste time on diagnosis and re-prompting. A model that finishes reliably shifts developer effort from restarting runs to code review, making agents delegatable rather than supervisory.
Implementation verdict
Replaces Sonnet 4.6 as the default model tier on GitLab Duo Agent Platform for everyday dev work. Requires GitLab Premium/Ultimate or paid credits; available now across all deployment models. Worth switching today if you're running agents in production—reliability improvements compound with cost efficiency.
Sources
Dev Signal
Get briefs like this in your inbox — free, every weekday.
100+ sources compressed into one 4-minute read. Ranked, cited, implementation-ready.