GitHub Copilot introduces three tiers for automatic model selection
GitHub Copilot is adding three tiers to automatic model selection – efficiency, balance and intelligence – for choosing between lower cost/speed and higher quality. They are gradually rolling out to VS Code, Copilot CLI and the Copilot app.
GitHub Copilot is introducing three new tiers for automatic model selection (auto model selection): efficiency, balance and intelligence. According to GitHub, the efficiency tier optimizes selection for lower cost and response speed, the intelligence tier selects the best available model based on output quality, and the balance tier seeks a compromise between cost and quality. All three tiers select a model from the same set of available models – only the selection criterion differs, not the model offering itself.
According to the company, the feature is gradually rolling out (rollout) to VS Code, Copilot CLI and the Copilot app. It was announced as part of the weekly GitHub Copilot update from 14 September 2026, which also included other, separate product changes.
The source article about the tiers does not provide specific prices, default settings or a list of models included in each tier. Details can be found in the source article.
Why it matters
Until now, GitHub Copilot users had to manage the trade-off between model cost, speed and quality by manually selecting a specific model; the new tiers simplify this to choosing a general focus (efficiency/balance/intelligence), with the system selecting a specific model from the same offering. This may make everyday work easier for individual developers and cost management easier for teams and organizations paying for Copilot usage.
Two audiences, two different impacts
What this means
For individuals
Developers using GitHub Copilot can now choose whether automatic model selection should prioritize lower cost and speed, the highest quality, or a balanced compromise between the two.
For a business
Companies using GitHub Copilot can indirectly influence the trade-off between model usage costs and output quality/speed by choosing a tier, because the efficiency tier targets lower cost and speed, while the intelligence tier targets quality.
ProductivityCheck the original
Event sources
clearly official source · 1 publisher, 0 independent. We count feeds from the same owner only once.