Changing the AI model for Copilot code completion

Learn how to change the default LLM for Copilot code completion to a different model.

In this article

Overview

By default, Copilot code completion uses the GPT-4o Copilot, a fine-tuned GPT-4o mini based large language model (LLM). This model has been trained on a wide range of high quality public GitHub repositories, providing coverage of over 30 programming languages. Its knowledge base is more current than the default model and you may find that it generates completion suggestions more quickly.

View the list of programming languages and technologies included in the training data.
  • C
  • C#
  • C++
  • Clojure
  • CSS
  • Dart
  • Dockerfile
  • Elixir
  • Emacs Lisp
  • Go
  • Haskell
  • HTML
  • Java
  • JavaScript
  • Julia
  • Jupyter Notebook
  • Kotlin
  • Lua
  • MATLAB
  • Objective-C
  • Perl
  • PHP
  • PowerShell
  • Python
  • R
  • Ruby
  • Rust
  • Scala
  • Shell
  • Swift
  • TeX
  • TypeScript
  • Vue

Note

The list of available models will change over time. When only one code completion model is available, the model picker will only show that model. Preview models and additional code completion models will appear in the picker when we release them.

You can switch AI models in the latest releases of VS Code with the latest version of the GitHub Copilot extension.

Effects of switching the AI model

Changing the model that's used for Copilot code completion does not affect the model that's used by Copilot Chat. See Changing the AI model for Copilot Chat.

There are no changes to the data collection and usage policy if you change the AI model.

If you are on a Copilot Free plan, all completions count against your completions quota regardless of the model used. See Plans for GitHub Copilot.

The setting to enable or disable suggestions that match public code are applied irrespective of which model you choose. See Finding public code that matches GitHub Copilot suggestions.

Enabling the model switcher

If you're using a Copilot Business or Copilot Enterprise plan, the organization or enterprise that provides your plan must enable the Editor preview features setting. See Managing policies for Copilot in your organization or Managing policies and features for Copilot in your enterprise.

Changing the AI model for code completion

The following instructions are for VS Code. If you are using Visual Studio, or a JetBrains IDE, click the appropriate tab at the start of this article.

  1. Open the command palette by pressing Ctrl+Shift+P (Windows/Linux) / Command+Shift+P (Mac).
  2. Type change completions model and select the "GitHub Copilot: Change Completions Model" command.
  3. In the dropdown menu, select the model you want to use.

Alternatively, if Command Center is enabled, you can click beside the icon at the top of the VS Code window, then click Configure Code Completions in the dropdown menu. Then choose Change Completions Model in the dropdown menu and select the model you want to use.

Checking which model is being used

  1. Open the Settings editor by pressing Ctrl+, (Linux/Windows) / Command+, (Mac).

  2. Type copilot completion and look for the "GitHub > Copilot: Selected Completion Model" section.

    The field in this section displays the currently selected model. If the field is empty, the default model is being used.