GitHub: Copilot Retires Opus 4.6 (fast) on June 29, Adds AGENTS.md to Code Review and ai_credits_used Field to API
GitHub announced three changes for Copilot through its changelog. The Opus 4.6 (fast) model is being retired on June 29, 2026 across all Copilot surfaces with a recommendation to switch to Opus 4.8 (fast), Copilot code review now reads repository AGENTS.md files, and the Copilot Usage Metrics API gains the ai_credits_used field tracking AI credit consumption per user.
This article was generated using artificial intelligence from primary sources.
GitHub announced three changes for Copilot, its AI programming tool, through its changelog. The changelog is a chronological record of minor product changes, separate from major blog posts, gathering practical updates for the day-to-day work of development teams.
Retiring the Opus 4.6 (fast) Model
The most important change for users is that the Opus 4.6 (fast) model is being retired on June 29, 2026 across all Copilot surfaces, with a recommendation to switch to Opus 4.8 (fast). The retirement affects chat, inline edits and code completion, so teams that have pinned the old model need to migrate before the deadline to avoid disruption.
AGENTS.md and Credit Measurement
Copilot code review now reads repository AGENTS.md files and adapts feedback to project conventions, reducing noise from irrelevant comments. Additionally, the Copilot Usage Metrics API gains a new ai_credits_used field that tracks total AI credit consumption per user in daily and 28-day reports. Unlike a billing statement, the field is a signal for consumption analysis and currently has no breakdown by feature, model or surface.
Frequently Asked Questions
- When is Opus 4.6 (fast) retiring in Copilot?
- The model is retiring on June 29, 2026 across all Copilot surfaces, with a recommendation to switch to Opus 4.8 (fast); the change affects chat, inline edits and code completion.
- What does AGENTS.md support bring?
- Copilot code review now reads repository AGENTS.md files and adapts feedback to project conventions.
- What does the ai_credits_used field measure?
- The field tracks total AI credit consumption per user in daily and 28-day reports, but without breakdown by feature or model yet.