GitHub Copilot code review switches from Lite to Balanced today
From 28 September the default effort level for Copilot code review moves to a higher-reasoning model, unless you already pinned your repo or org to Lite.
GitHub’s own review bot got more cautious today. As of 28 September, the default effort level for Copilot code review switches from Lite to Balanced, for both existing and new repositories and organisations, according to GitHub’s changelog.
Lite gives focused feedback on straightforward changes such as documentation updates and small fixes. Balanced uses what GitHub calls “a higher-reasoning model” to dig into complex logic, security-sensitive code and cross-service changes. Anyone who wanted to keep Lite needed to set it explicitly in their repository or organisation settings before today; leaving the setting on Default now gets Balanced. The change sits inside a wider 28 August announcement that also moves Copilot Chat on github.com to retain conversation history for the life of an account instead of 28 days, once a “unified Copilot experience” rolls out no earlier than today.
Why it matters: if your team relies on Copilot’s automatic pull request reviews, expect slower, more thorough comments from now on rather than the quick pass Lite gave you. That trade makes sense for anything touching auth, payments or cross-service logic, but if you were counting on Lite’s speed for routine PRs, check your org’s Copilot settings now rather than after review latency turns into a complaint.
The caveat: GitHub hasn’t published a pricing difference between the two effort levels, so it isn’t clear yet whether Balanced reviews draw more from your plan’s usage than Lite did.