Continue

Continue (continue.dev) is an open-source AI coding assistant released under the Apache 2.0 license. It installs as an extension in VS Code or JetBrains IDEs and provides three core capabilities: autocomplete (inline suggestions as you type), chat (contextual conversation about your code), and edit (apply AI changes directly to selected code).

What distinguishes Continue from proprietary alternatives like GitHub Copilot is its model flexibility and privacy model. Continue works with any LLM: Claude via API, GPT-4 via OpenAI API, Gemini, local models via Ollama (Llama, Mistral, CodeLlama), and dozens of others. Organizations can route code completions to a local model for privacy while using a cloud model for chat — or run everything locally with no external calls at all.

Context retrieval is a core engineering focus. Continue uses codebase indexing to find relevant files, functions, and documentation before constructing prompts. Users can explicitly add context with @file, @folder, @docs, @terminal, @codebase, and @web references. The assistant sees what the developer sees: the current file, recent edits, terminal output, and documentation.

The extension is actively maintained with over 10,000 GitHub stars and weekly releases. The project raised venture funding in 2024 to build enterprise features including team context sharing, usage analytics, and centralized model configuration for organizations.

For individual developers, Continue is entirely free with no usage limits beyond what the underlying LLM API charges. Enterprise pricing is available for team features.

The project is designed for developers who want AI assistance without the lock-in, data privacy tradeoffs, or model constraints of closed alternatives.