Copilot code review reports quota limit while premium credits remain #208758
Replies: 2 comments
|
Thank you for your interest in contributing to our community! We currently only accept discussions created through the GitHub UI using our provided discussion templates. Please re-submit your discussion by navigating to the appropriate category and using the template provided. This discussion has been closed because it was not submitted through the expected format. If you believe this was a mistake, please reach out to the maintainers. |
|
Uh oh!
There was an error while loading. Please reload this page.
🏷️ Discussion Type
Bug
💬 Feature/Topic Area
Copilot in GitHub
Body
Copilot code review refuses with a quota-limit message even though the same account still has premium credits remaining. The refusal does not consume those credits.
Account
copilot_plan=individual,access_type_sku=free_educational_quotaoverage_permitted=falseWhat happened
On yyjeqhc/webcodex#666 I requested a Copilot review twice. Both reviews are from
copilot-pull-request-reviewerand contain only:There are no inline review comments. The pull request is 25 files, +938 / −28. The timeline shows both requests were made by ITSGJY, so the message is about this account, not the repository owner.
Quota after the refusals
At 2026-09-24T22:42:00Z, after both refusals,
GET /copilot_internal/userfor this account still reported:premium_interactions.entitlement: 200premium_interactions.credits_used: 127premium_interactions.quota_remaining: 72.6premium_interactions.percent_remaining: 36.3premium_interactions.remaining: 72premium_interactions.has_quota: truepremium_interactions.unlimited: falsepremium_interactions.overage_count: 0chatandcompletions: unlimitedSeptember billing usage for this account lists Copilot AI Credits quantity 127.36, which matches
credits_used, not an exhausted allowance. The two refusals did not reducequota_remaining.That payload only has
chat,completions, andpremium_interactions. There is no separate code-review counter. The code-review documentation says each review spends AI credits, a manual request is billed to the person who asked, and exhausting a credit budget blocks code review together with other credit-consuming features. That does not match a refusal while 72.6 credits remain and the other counters are untouched.Expected
Either the review runs and deducts credits, or the quota message names a counter that is actually exhausted. Here the review service and the account usage snapshot disagree.
All reactions