Continue

Open-source IDE extension for building your own AI code assistant

AI Coding & DevelopmentFree planOverseasβ˜…β˜…β˜…β˜…β˜† 4.0

What is Continue?

Continue is an open-source extension for VS Code and JetBrains IDEs that turns chat, inline edits and tab autocomplete into a configurable assistant. Rather than shipping one fixed model, it reads a config file where you declare which model handles chat, which handles completion, and which context providers feed the prompt: open files, documentation, terminal output or a repository index. That makes it a common choice for teams that must route inference through a local model or a self-hosted endpoint for data-residency reasons. Rules files let you encode project conventions so suggestions follow the codebase instead of a generic style.

Last updated: 2026-09-20. This site only provides an index; for exact features, pricing, and licensing, see the official website.

Key features

  • Chat, inline edit and tab autocomplete in a single extension
  • Model-agnostic config for hosted APIs or local models via Ollama
  • Context providers for files, docs, terminal output and retrieval
  • Rules and prompt files that encode project conventions
  • Ships for VS Code and JetBrains IDEs from one shared config
  • Open source, with a hub for sharing reusable assistant blocks

Pros & cons

Strengths

  • Free and open source, with no vendor lock-in on models
  • Local model support suits strict data-residency requirements
  • Config lives in the repository, so a team can review it in pull requests

Watch out for

  • Setup is more involved than a one-click commercial assistant
  • Answer quality depends heavily on the model you connect
  • JetBrains support tends to lag behind the VS Code extension

Best for & use cases

self-hosted ai assistants, local model setups, data residency rules and custom code assistants

If you're comparing similar products, check the alternatives below, or browse all tools in the AI Coding & Development category.

FAQ

Do I need my own API key?

Not necessarily. Continue works with hosted providers using a key you supply, and it can also point at a local model served by Ollama or another runtime, which is common when code must not leave the machine.

Which editors are supported?

VS Code and the JetBrains IDEs are the primary targets and share one configuration format. Feature parity is closest in VS Code, and JetBrains releases tend to follow a little behind.

How is it different from GitHub Copilot?

Copilot is a managed service tied to GitHub and a fixed set of models. Continue is a client you configure, so you pick the models, the context and where inference runs, including fully local setups.