Continue is the open-source AI code assistant that puts you back in the driver's seat. Instead of being locked into one vendor's models and one way of working, Continue lets you plug in any LLM - local or cloud - and customize how AI shows up in your IDE. It's like building your own GitHub Copilot but with the freedom to use local models for privacy, cloud models for power, or both, all without a monthly subscription dictated by someone else's roadmap.

Main Features

• Bring your own model: Connect to OpenAI, Anthropic, Google, Mistral, Ollama, LM Studio, or any OpenAI-compatible endpoint - you choose the brain.
• Tab autocomplete: Context-aware code completions that work with any configured model, including free local models running entirely on your machine.
• Chat with your codebase: Ask questions, get explanations, request refactors, and generate code with @mentions for files, folders, and documentation.
• Fully open source (Apache 2.0): Every line of code is public and auditable - no black boxes, no hidden prompts, no vendor lock-in.
• Custom slash commands: Create reusable prompt templates like /explain, /refactor, or /generate-tests that fit your team's exact workflow.
• VS Code and JetBrains support: Runs in VS Code (where it started) and JetBrains IDEs with consistent behavior across both platforms.
• Local-first and offline capable: Run models entirely on your machine via Ollama - no internet required, no code leaves your device.
• Config-as-code (.continuerc.json): Share AI assistant configuration across your team via version control - everyone gets the same models, prompts, and behavior.

Who Should Use It?

• Privacy-conscious developers: At healthcare, finance, or defense companies who need code AI that never sends code to a third party.
• Open-source enthusiasts: Who want full control over their AI tools and the ability to inspect, modify, and contribute.
• Developers in air-gapped environments: Working on classified or proprietary systems where cloud-based AI tools are simply not allowed.
• Teams wanting consistency: Who can share model config and slash commands via git so every developer has the same AI experience.
• Cost-conscious engineers: Who want free local models for routine tasks and cloud models for complex problems - without two separate subscriptions.
• VS Code and JetBrains users: Who want an AI assistant that works exactly the same way regardless of which IDE they're in.