Continue
An open-source IDE extension for building custom AI coding assistants.
Continue is an open-source extension for VS Code and JetBrains that lets you assemble your own AI coding assistant. It supplies the editor integration — chat, tab autocomplete, inline edits, and agent modes — while you pick the model: hosted providers like Anthropic and OpenAI, or local models via Ollama, all in shareable, version-controlled config.
Continue is an open-source extension for VS Code and JetBrains IDEs that lets developers assemble their own AI coding assistant. Instead of locking you into one vendor's model, it provides the editor integration (chat, autocomplete, inline edits, and agent actions) while leaving the choice of model and provider up to you.
It is aimed at developers and teams who want control over which LLMs they use, where requests are sent, and how the assistant is configured. Because the configuration lives in your project as code, setups can be version-controlled and shared across a team.
Highlights
- Works in VS Code, JetBrains IDEs, and a terminal CLI
- Bring-your-own model: connect hosted providers (OpenAI, Anthropic, etc.) or local models via Ollama and similar runtimes
- Chat, tab autocomplete, inline edit, and agent modes
- Context providers that pull in files, docs, terminal output, and codebase search
- Declarative configuration so assistants are reproducible and shareable
A typical workflow connects a model in a config file, then uses chat for questions, autocomplete while typing, and inline edits for targeted refactors. A minimal model entry looks like this:
models:
- name: my-assistant
provider: anthropic
model: claude-sonnet-4-6
apiKey: ${ANTHROPIC_API_KEY}Good to know
NOTE
Continue is open-source (Apache 2.0) and free to install, but you supply your own model access. Running hosted models incurs provider API costs; local models avoid that but require capable hardware.
Frequently asked questions
- What is Continue?
- Continue is an open-source extension for VS Code and JetBrains IDEs (plus a terminal CLI) that lets developers assemble their own AI coding assistant. It provides chat, tab autocomplete, inline edit, and agent modes, while leaving the choice of model and provider up to you — and because configuration lives as code, setups can be version-controlled and shared across a team.
- Is Continue free?
- Yes — Continue is open source under Apache 2.0 and free to install. You supply your own model access: hosted models incur provider API costs, while local models via Ollama and similar runtimes avoid that but require capable hardware.
- How do I use Continue with my own model?
- Connect a model in a config file, then use chat for questions, autocomplete while typing, and inline edits for targeted refactors. A minimal model entry declares a name, a provider (e.g. anthropic), a model (e.g. claude-sonnet-4-6), and an apiKey reference like ${ANTHROPIC_API_KEY}.
Related
- AI Coding Agents in 2026: The Open-Source & CLI EditionCursor and Windsurf vs the open-source agents — OpenCode, Cline, Aider, Codex CLI, and more. Who should bring their own model, and when to stay in the terminal.
- CodySourcegraph's AI coding assistant for the IDE, grounded in deep codebase context.
- Kilo CodeOpen-source AI coding agent extension for VS Code and JetBrains, built as a superset of Roo Code and Cline, with bring-your-own-key and zero model markup.
- TabbySelf-hosted, open-source AI coding assistant by TabbyML — run your own completion and chat models on your infrastructure, with IDE extensions.
- TabnineAn AI code completion and chat assistant built around code privacy, self-hosting, and air-gapped enterprise deployment.