GitHub Copilot Auto Model Selection in JetBrains
GitHub Copilot now supports auto model selection in JetBrains IDEs, streamlining the coding process for vision and deep learning projects. This means less time troubleshooting dependencies or model versioning, and more time focusing on creative application and experimentation. Documenting the use of Copilot can show how these tools accelerated development and helped avoid common pitfalls.
GitHub Copilot's auto model selection is now generally available in JetBrains IDEs for all Copilot plans. This feature allows Copilot to dynamically choose the best available model based on real-time system health and model performance, reducing rate limiting and latency. Copilot intelligently routes to models like GPT-5.4, GPT-5.3-Codex, Sonnet 4.6, and Haiku 4.5 depending on the user's plan and policies. The "auto" setting gives users reliable access to their preferred models without manual switching. Users can see which model was used by hovering over the model response and can switch between "auto" and any specific model at any time. Administrators' model settings are always respected. For paid plans using auto model selection, a 10% multiplier discount is applied. This means that when "auto" uses a model with a 1x multiplier, users will draw down 0.9 premium requests instead of 1. Soon, Copilot will be able to select the most appropriate model for the task, matching the model to the complexity level of the request.
