Gemini Code Assist overview

Gemini Code Assist, which is a product in the Gemini for Google Cloud portfolio, offers AI-powered assistance to help your development team build, deploy, and operate applications throughout the software development lifecycle.

You can use Gemini Code Assist in your IDE (such as VS Code, JetBrains IDEs (such as IntelliJ and PyCharm), Cloud Workstations, or Cloud Shell Editor) for AI-powered coding assistance in many popular languages. You can get code completions as you write your code, generate full functions or code blocks from comments, generate unit tests, and get help with debugging, understanding, and documenting your code.

Gemini Code Assist provides contextualized responses to your prompts, including source citations regarding which documentation and code samples Gemini used to generate its responses.

The Gemini large language models (LLMs) that are used by Gemini Code Assist are trained on datasets of publicly available code, Google Cloud-specific material, and other relevant technical information in addition to the datasets used to train the Gemini foundation models. Models are trained so that Gemini Code Assist responses are as useful to Gemini Code Assist users as possible.

Learn how and when Gemini for Google Cloud uses your data. As an early-stage technology, Gemini for Google Cloud products can generate output that seems plausible but is factually incorrect. We recommend that you validate all output from Gemini for Google Cloud products before you use it. For more information, see Gemini for Google Cloud and responsible AI.

How Gemini Code Assist helps with coding

Gemini Code Assist provides assistance with coding and code-related tasks. We recommend that you test and review all code. Gemini Code Assist provides citation information when it directly quotes at length from another source, such as existing open source code. For more information, see How and when Gemini cites sources.

Interact with Gemini Code Assist in your IDE

After you set up Gemini Code Assist for a Google Cloud project, and install the Gemini Code Assist extension in your IDE (VS Code or supported JetBrains IDE), you can ask for assistance in the following ways:

  • Receive code completions or generate code directly in the code editor.

  • Click spark Gemini in the IDE to display the conversational assistant. You can ask questions or select code in your editor and enter prompts such as the following:

    • Write unit tests for my code.
    • Help me debug my code.
    • Make my code more readable.

For more information, see Use Gemini Code Assist in your IDE.

Supported features

The following sections show the types of generative AI assistance that are available in each Gemini Code Assist edition.

Coding assistance and chat

The following table shows the types of generative AI assistance that are available in supported IDEs:

AI coding assistance Gemini Code Assist Standard Gemini Code Assist Enterprise

Code completion and generation in your IDE project in the following IDEs:

Conversational assistant in your IDE using your opened files' context
Multi-IDE support (VSCode, JetBrains IDEs such as IntelliJ and PyCharm, Cloud Workstations)
Smart actions and commands
Initiate smart actions by right-clicking selected code (VS Code, JetBrains IDEs such as IntelliJ and PyCharm, Cloud Shell, and Cloud Workstations). Initiate smart commands with the slash / on the quick pick bar either with or without selected code (VS Code, Cloud Shell, and Cloud Workstations).
Intellectual property and compliance
Source citations in your IDE and the Google Cloud console
IP indemnification
VPC-SC and Private Google Access
Enterprise knowledge
Customized code suggestions from your code bases in GitHub and GitLab in your IDE  

Gemini in Apigee

The following table shows the types of generative AI assistance with API development in Apigee (IDE and the Google Cloud console):

Gemini Code Assist for API management Gemini Code Assist Standard Gemini Code Assist Enterprise
Enterprise context used when creating API specifications. (Preview)  
Smart Search powered by Vertex AI in API hub. (GA)  
Gemini Code Assist code explained for Apigee policies. (Preview)  

Gemini in Application Integration

The following table shows the types of generative AI assistance in Application Integration in the Google Cloud console:

Gemini in BigQuery features with Gemini Code Assist

The following table shows the types of generative AI assistance for BigQuery in BigQuery Studio:

Analytics code assist Gemini Code Assist Standard Gemini Code Assist Enterprise
SQL code assistance with generation, completion (preview), and explanation  
Python code assistance with generation and completion (preview)  
Data to insights
Data canvas to find, transform, query, and visualize data with natural language prompts and a UI for analysis workflows  
Data insights library of prevalidated SQL questions for data assets  

Gemini in Colab Enterprise

The following table shows the types of generative AI assistance for code in Colab Enterprise:

Notebook code assist Gemini Code Assist Standard Gemini Code Assist Enterprise
Python code generation and completion in notebook

Gemini in Databases

The following table shows the types of generative AI assistance for coding in databases, including Database Studio:

Generate SQL queries (Database Studio) Gemini Code Assist Standard Gemini Code Assist Enterprise
Write in natural language to generate SQL statements.
Get contextual code that works with your schema.
Optimize and explain existing queries.

Gemini in Firebase

The following table shows the types of generative AI assistance for application development provided by Gemini in Firebase and AI assistance in Crashlytics available in the Firebase console:

Coding and product AI assistance and chat Gemini Code Assist Standard Gemini Code Assist Enterprise
Generate, refactor, and debug sample code for Firebase with natural language in chat.
Use natural language prompts to explain, generate, and transform code in the Gemini pane.
Use deep knowledge, best practices, and troubleshooting expertise for Firebase products and services.
App quality analysis
Summarize app crashes and provide insights and troubleshooting steps to help developers investigate and resolve app quality issues.
Analyze existing code, identify potential issues, and suggest improvements.
Contextual awareness
Use project and application context to guide conversational assistance, troubleshooting, and app quality analysis.

Set up Gemini Code Assist

For detailed setup steps, see Set up Gemini Code Assist.

What's next