Continue

What is Continue about?

Continue provides an AI‑powered development experience that lives inside your favorite editors (VS Code, JetBrains IDEs) and the command line. It brings a conversational agent, inline code suggestions, and one‑click edits directly to the context you are working in, turning AI into a collaborative teammate.

How to use Continue?

1. IDE installation – Add the extension from the VS Code Marketplace or the JetBrains plugin repository. The extension activates automatically and prompts you to configure your LLM provider.
2. CLI usage – Run the tool without a global install:

   npx @continue/cli init
   

   The CLI lets you start a chat session, run edits, or invoke custom agents from any terminal.
3. Configuration – After installation, open the settings panel to select a model (e.g., Claude, OpenAI, Qwen) and provide the required API key. Once configured, you can invoke the Agent, Chat, Edit, or Autocomplete features via shortcuts or command palette.

Key features of Continue

• Agent – An interactive AI that can take on development tasks, answer questions, and guide you through problem‑solving.
• Chat – Free‑form conversational interface for code queries, documentation lookup, and brainstorming.
• Edit – Generate precise code modifications without leaving the current file; the AI returns a diff that can be applied instantly.
• Autocomplete – Inline suggestions appear as you type, powered by large language models.
• Cross‑environment – Works in VS Code, JetBrains IDEs, and the terminal, and can be scripted into CI pipelines.
• Open‑source – Licensed under Apache 2.0, allowing community contributions and self‑hosting.

Use cases of Continue

• Rapid prototyping: generate boilerplate and iterate on functions directly in the editor.
• Debugging assistance: ask the Agent to explain stack traces or suggest fixes.
• Refactoring: use Edit to apply consistent code style changes across a project.
• Documentation: generate docstrings, READMEs, or API docs on demand.
• CI automation: embed the CLI in CI jobs to run code reviews or automated refactors before merge.

FAQ from Continue

Is Continue free? The core extension and CLI are free and open‑source. Using certain LLM providers may incur their own API costs.

Which models are supported? Claude, OpenAI GPT‑4/3.5, Anthropic, Qwen, and any OpenAI‑compatible endpoint can be configured.

Can I run Continue without an internet connection? Self‑hosting a compatible LLM locally is possible, but the default setup requires internet access to query the model APIs.

How does Continue handle privacy? All code snippets are sent only to the configured model endpoint. No telemetry is sent to Continue servers unless you opt‑in.

Is there a way to create custom agents? Yes – you can define custom prompts and workflows in the continue.config.js file and invoke them via the CLI or editor commands.

Ship faster with Continuous AI

Build and run custom agents across your IDE, terminal, and CI

Get started in VS Code, JetBrains, or CLI

Agent

Agent to work on development tasks together with AI

Chat

Chat to ask general questions and clarify code sections

Edit

Edit to modify a code section without leaving your current file

Autocomplete

Autocomplete to receive inline code suggestions as you type

Contributing

Read the contributing guide, and join #contribute on Discord.

License

Apache 2.0 © 2023-2024 Continue Dev, Inc.