Continue is an open-source AI coding assistant that integrates directly with VS Code, JetBrains, Cursor, and other IDEs. It acts as a customizable alternative to tools like GitHub Copilot, letting you bring your own LLM provider (OpenAI, Anthropic, Meta, self-hosted models) while keeping full control over how AI interacts with your codebase.
Most AI coding tools lock you into a specific model or cloud environment. Continue gives developers flexibility: you choose the model, the context window, and whether your code ever leaves your machine. It helps teams who care about privacy, cost control, or on-premise deployment. Solo developers benefit from a strong autocomplete and “edit this code” workflow without committing to a subscription.
Continue adds a sidebar and in-editor commands that let you:
- Ask the model questions about your codebase.
- Generate new code or explain existing code.
- Run “edit commands” that rewrite selections with natural language prompts.
- Use RAG-style context injection to pull in relevant files, diffs, documentation, or repos.
- Connect to external APIs or self-hosted LLMs, including via Ollama or LM Studio.
It’s highly configurable: you can define your own models, prompt templates, context rules, and shortcuts. It supports large-context models so it can analyze multi-file refactors or entire repositories.
