AI BriefingGitHub Copilot Rolls Out Tiers of Auto for Cost and Quality Control
Key point
GitHub Copilot has made Tiers of Auto available to all customers, allowing developers to select Efficiency, Balance, or Intelligence modes in VS Code, CLI, and the Copilot App to optimize model routing.
Details
GitHub Copilot has rolled out Tiers of Auto, a feature allowing developers to manually select cost and quality tradeoffs within the model picker. Available across VS Code, CLI, and the Copilot App, the update introduces three distinct optimization modes:
- Efficiency: Prioritizes low costs for fast, straightforward tasks.
- Balance: Weighs cost, quality, and latency for general use.
- Intelligence: Prioritizes quality for complex tasks.
Model Routing and Updates
This release follows the September 14, 2026 announcement and marks a shift toward greater developer control in intelligent model routing. The update includes new models in the auto-mix, such as GPT 6 Astra, the GPT 5.6 series, and Opus 5. Additionally, Adaptive multi-turn support now allows Auto to reevaluate complexity throughout a session, rather than only at the start or after compaction, improving cache preservation and cost efficiency.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.