How to Set Up Continue.dev with Ollama in VS Code on Linux

Why Continue.dev + Ollama?

Cloud AI coding assistants can be expensive and may raise data compliance concerns for proprietary codebases. Continue is an open-source VS Code extension that connects directly to local inference backends like Ollama, giving you free, private code completion.

Step 1: Pull Recommended Coding Models

For chat, inline refactoring, and code generation, the Qwen 2.5 Coder family offers state-of-the-art capability:

# Main chat and refactoring model
ollama pull qwen2.5-coder:7b

# Fast low-latency model for tab autocomplete
ollama pull qwen2.5-coder:1.5b

Step 2: Installing the Continue Extension

Open VS Code on your Linux machine, press Ctrl+P, and enter:

ext install Continue.continue

Step 3: Configuring Local Model Endpoints

Click the gear icon in the bottom right of the Continue sidebar to open ~/.continue/config.yaml. Update your configuration to use local Ollama models:

models:
  - name: "Qwen 2.5 Coder 7B"
    provider: "ollama"
    model: "qwen2.5-coder:7b"

tabAutocompleteModel:
  title: "Qwen Coder 1.5B"
  provider: "ollama"
  model: "qwen2.5-coder:1.5b"

Step 4: Using Keyboard Shortcuts

Test the integration using Continue's primary keybindings:

• Ctrl+L: Open the Continue chat pane with your selected code block attached.
• Ctrl+I: Inline code refactor prompt directly inside the editor buffer.
• Tab: Accept real-time code completions as you type.

You now have a fully functional local AI pair programmer running on Ubuntu without monthly subscription fees.