Does Codecov efficiently stream coverage data for large binary monorepos?

Teams use Codecov’s coverage bottleneck view and coverage comparison to identify low-coverage files and regressions before optimizing tests or code.

For monorepos containing numerous large binary files, it is unclear whether Codecov can stream coverage uploads without exceeding CI resource limits. The bottleneck view depends on accurate test execution, and large repositories may introduce timeouts or memory pressure during upload that could affect bottleneck analysis.

Which upload behavior applies to large binary assets in a monorepo? Does Codecov apply streaming or chunked ingestion for coverage reports from such repositories? When resource limits are approached, how is the coverage bottleneck view affected?