GitHub Copilot adds tiers for cost and quality in auto model selection
GitHub Copilot's auto model selection now offers three configurable tiers: efficiency, balance, and intelligence. These tiers allow users to prioritize cost, quality, and response time for each prompt, with Copilot optimizing model selection accordingly. The feature is currently rolling out across Visual Studio Code, Copilot CLI, and the GitHub Copilot app. Usage is billed based on the chosen model, with paid subscribers retaining a 10% discount on auto-selected models.
Features (1) ›
- Configure cost and quality in Copilot auto model selection
GitHub Copilot auto model selection now provides three tiers—efficiency, balance, and intelligence—allowing users to specify how cost, quality, and response time are weighed for each prompt. The 'Efficiency' tier prioritizes low costs, 'Balance' offers a mix, and 'Intelligence' focuses on quality for complex tasks. This functionality is rolling out in Visual Studio Code, Copilot CLI, and the GitHub Copilot app.
https://github.blog/changelog/2026-09-14-configure-cost-and-quality-in-copilot-auto-model-selection
Related releases
- GitHub Copilot usage metrics now include VS Code Agents activity GitHub Changelog ·
- GitHub Copilot Code Review Gains Auto-Resolution, Smart Commits, and Deeper Analysis GitHub Changelog ·
- GitHub Copilot Weekly Releases: Jira Integration, HydraFusion, and Enterprise Controls GitHub Changelog ·
- GitHub Launches Refreshed Repository Pull Requests Page in Public Preview GitHub Changelog ·
- GitHub AI Scan for PRs: API Management Now in Public Preview GitHub Changelog ·
- Control GitHub Actions Cache Access with cache-mode GitHub Changelog ·