Gemini Code Assist

Gemini Code Assist is Google's AI-powered coding assistant, built on the Gemini family of large language models and designed to accelerate software development across the entire lifecycle. Originally launched as Duet AI for Developers and rebranded in 2024, it provides intelligent code completion, natural-language code generation, chat-based assistance, code transformation, and automated code review inside popular IDEs and across Google Cloud surfaces. Developers can use it in Visual Studio Code, JetBrains IDEs (IntelliJ, PyCharm, GoLand, WebStorm, Android Studio), Cloud Shell Editor, Cloud Workstations, BigQuery, Firebase Studio, Apigee, Application Integration, and the Google Cloud Console, making it one of the most broadly integrated AI coding tools in the enterprise ecosystem.

The product is offered in multiple tiers. Gemini Code Assist for Individuals is a free plan aimed at students, hobbyists, and individual developers, offering generous monthly quotas for code completions and chat interactions. Gemini Code Assist Standard and Enterprise are paid tiers intended for professional teams, adding higher quotas, enterprise-grade security, IP indemnification, Google Cloud service integrations, and — at the Enterprise tier — codebase awareness that lets the assistant reason over private repositories connected via GitHub, GitLab, Bitbucket, or Google-hosted sources. A separate Gemini Code Assist for GitHub app brings AI-powered pull request reviews and summaries directly into GitHub workflows.

Under the hood, Gemini Code Assist uses Google's Gemini 2.5 models with a large context window (up to 1M tokens on enterprise tiers), enabling it to reason about long files, multi-file changes, and substantial portions of a codebase at once. It supports more than 20 programming languages including Python, Java, JavaScript/TypeScript, Go, C/C++, C#, Kotlin, Rust, PHP, Ruby, Scala, SQL, and Terraform/HCL. Beyond completions, it offers inline code transformations (explain, refactor, fix, generate tests, document), agentic workflows for multi-step tasks, customization through private-code context, and deep integration with Google Cloud services so developers can query BigQuery, scaffold Firebase apps, generate Apigee API specs, or troubleshoot Cloud logs in natural language. For enterprises, it provides admin controls, audit logs, data residency options, and a contractual guarantee that customer prompts and code are not used to train Google's foundation models.