Skip to main content
The Cerebras integration for VS Code allows you to use Cerebras’ fast inference for AI-powered coding assistance directly within the editor.

​
Prerequisites

Before you begin, ensure you have:
  • Cerebras API Key - Get a free API key here.
  • VS Code - Download and install from code.visualstudio.com. Gemma 4 31B support requires VS Code 1.120 or later and Cerebras extension 0.1.21 or later.
  • GitHub account that is not enrolled in the GitHub Copilot Enterprise plan.
step1

​
Configure VS Code

1

Sign up for a free GitHub Copilot account

Follow the instructions in VS Code to sign up for a free GitHub Copilot account. You can also click on the Extensions tab on the left side and search for Copilot.step2
2

Install the Cerebras extension

Click on the Extensions tab on the left side. Search for “Cerebras” and click Install.step3
3

Configure API key

At the bottom of the chat window, click on the model selection list (this is set to Auto by default), then click Other Models.step4Click on Manage Models.step5Click +Add Models and choose Cerebras from the list.step6Enter your Cerebras API key.step7
4

Configure Models

Select which models to enable by clicking the eye icon next to each one.step8
5

Select a model and start coding!

step9

​
Available Models

The Cerebras VS Code extension supports these models:
Gemma 4 31B requires Cerebras extension 0.1.21 or later. Free-tier requests support up to 65K input tokens and 32K output tokens. Paid requests support up to 131K input tokens and 40K output tokens.
See the Gemma 4 31B model page for image requirements and current limits.