GitHub Copilot auto model selection now routes to the best model for your task in Copilot CLI, using utilization and model health metrics for a high quality, reliable, and token-efficient experience.
How it works Auto weighs real-time model availability and reliability signals, then evaluates your task across several dimensions like reasoning, code generation complexity, bug diagnosis difficulty, and tool orchestration needs to select the optimal model.
Stay in control: Switch between** Auto**and any specific model at any time with the/model
command.Respects your policies: Auto honors all model policies set by admins.** Diverse model access**: Autoleverages models from multiple model families, depending on subscription type and policies. Models will change over time.
Premium request use Auto is billed based on the model it selects, which is currently limited to models with 0x to 1x multipliers. Paid subscribers get a 10% discount on the model multiplier when using auto (e.g., when auto uses a model that has a 1x multiplier, you will draw down 0.9 premium requests instead of 1).
Getting more out of Copilot Auto routes along natural cache boundaries to avoid unnecessary cache related costs. Our evaluations show gains in token efficiencies with no quality regression, as not all tasks require a high reasoning or token-intensive model.
No setup is required. Update to the latest version of Copilot CLI and choose Auto to get started. For more information, see our documentation about auto model selection. Join the discussion within GitHub Community.