GitHub Copilot Code Review Default Effort Level

Did GitHub Copilot code review change its default effort level

Yes. GitHub changed the default review effort level for Copilot code review from Lite to Balanced. The current changelog says the default is changing from Lite to Balanced, and the newer documentation reflects Balanced as the current default for the effort-level system.

The part people get wrong is reading an older page and assuming it is still current. GitHub’s docs and changelog now sit next to each other, but they are not all updated at the same time. One docs page still says Lite is the default in the effort-level description, while the newer changelog announces the switch to Balanced, so the safest reading is that the default changed and some pages lagged behind.

GitHub also changed the naming around the same time. The newer docs and changelog use Lite and Balanced, while an older docs snapshot still shows Low and Medium. That is the inconvenient part for anyone comparing screenshots, configs, or cached docs, because the labels and the default can move together and make old instructions look current when they are not.

If you manage repositories or organizations, check the setting instead of assuming the platform default. GitHub lets repository administrators and organization owners set a default review effort level for automatic code reviews, so your local configuration can override the platform’s baseline. A repo can therefore behave differently from what the headline default says.

The practical difference matters. Lite is the faster, more targeted review, while Balanced does deeper analysis and uses more AI credits, with a possible increase in GitHub Actions usage. If your team expects the old default behavior, the review may feel slower or cost more credits after the change, even when nobody touched repository settings.

A simple way to verify your own setup is to open the repository or organization Copilot code review settings and look at the selected review effort level. GitHub documents that setting path directly, and it is the source of truth for how new automatic reviews run in that scope. If the setting says Balanced, that is what new automatic reviews use, regardless of what an older page or screenshot suggests.

One more detail matters if you are reading release notes for history. GitHub published an August 7, 2026 changelog entry specifically announcing that Lite and Balanced effort levels were generally available, and then later published an August 28, 2026 changelog note saying the default review effort level was changing from Lite to Balanced. That sequence confirms the default did change after the effort levels became generally available.

If you are writing docs or a reference page, use the current default carefully and name the source. The current state is not “Copilot review always uses Lite” and it is not “the default never changed.” The correct short answer is that it changed, and repository or organization settings can still override the baseline.

For teams comparing behavior across dates, the right test is to look at the pull request overview comment GitHub now emits, because GitHub says it labels which effort level ran for each review run. That gives you a concrete record of what actually happened on a given pull request, instead of guessing from documentation that may have been written before the default switched.