- Selection preferences do not set a fixed price.
What changed
GitHub’s September 14 changelog introduces efficiency, balance and intelligence tiers for Copilot’s automatic model selection. They prioritize different combinations of price, response time and quality. All use the same model pool, and Copilot still evaluates each prompt separately. A simple request can therefore reach a smaller model even with intelligence selected.
Why it matters
For developers, this adds a preference between manually choosing a model and accepting an undifferentiated automatic choice. Billing follows the model actually selected, so the tier itself does not establish a fixed task price.
What remains unproven
The rollout is continuing across three clients. GitHub supplies no comparative accuracy or latency measurements in this announcement, leaving the practical differences between tiers to be established.
Read beyond this page.
Recorded source-check date: 15 Sep 2026. A link is not, by itself, evidence that every claim has been independently verified.
Changes & version history
Version 3 · 15 Sep 2026
Scheduled release of checksum-bound AI-assisted editorial review
Version 2 · 15 Sep 2026
Checksum-bound editorial review scheduled for release
Version 1 · 15 Sep 2026
Source-linked private review edition