Per-model token breakdown in the usage report
GitHub's latest changelog entry announces a new feature in the usage report: a per-model token breakdown. Previously, the usage report likely aggregated token usage across all models, but now users can view token consumption for each model separately. This enhancement provides more granular visibility into how tokens are being used, which can help developers and organizations better understand their API spending and optimize their usage patterns. The update is documented in the GitHub Changelog, dated August 11, 2026, and is part of GitHub's ongoing improvements to its developer tools. The exact implementation details, such as how the breakdown is displayed or whether it applies to all API endpoints, are not specified in the source text. However, the core change is clear: the usage report now includes a per-model token breakdown, giving users more detailed data about their token usage.
Developers can now see token usage per model, enabling more precise cost tracking and optimization.