Continue

Overview

Continue is an open-source AI coding assistant that works as an extension for VS Code and JetBrains IDEs. It provides code completion, chat, and refactoring capabilities with support for multiple AI models. Continue emphasizes privacy and flexibility, allowing users to use their preferred AI provider or run models locally.

Continue has become popular for its open-source approach and flexibility in model selection, making it a great choice for privacy-conscious developers.

Features

• Open-source and free — No cost, fully transparent
• Multi-model support — Claude, GPT, Gemini, local models
• VS Code extension — Deep VS Code integration
• JetBrains extension — Also available for IntelliJ, PyCharm, etc.
• Privacy-focused — Local model support for complete privacy
• Customizable — Extensive configuration options

Installation

VS Code

1. Open VS Code
2. Go to Extensions (Ctrl+Shift+X)
3. Search for "Continue"
4. Click Install

JetBrains

1. Open your JetBrains IDE
2. Go to Settings > Plugins
3. Search for "Continue"
4. Click Install

Configuration

Edit .continue/config.json:

{
  "models": [
    {
      "title": "Claude",
      "provider": "anthropic",
      "apiKey": "your-api-key",
      "model": "claude-3-5-sonnet"
    },
    {
      "title": "Local",
      "provider": "ollama",
      "model": "codellama:7b"
    }
  ],
  "tabAutocompleteModel": {
    "title": "CodeLlama",
    "provider": "ollama",
    "model": "codellama:7b"
  }
}

Core Concepts

Model Flexibility

Continue supports many providers:

• Anthropic Claude — Best for complex reasoning
• OpenAI GPT — Wide compatibility
• Google Gemini — Good value
• Local models — Ollama, LM Studio, vLLM
• Azure OpenAI — Enterprise deployment

Privacy Options

For complete privacy:

{
  "models": [{
    "title": "Local CodeLlama",
    "provider": "ollama",
    "model": "codellama:7b"
  }]
}

Examples

Chat with Context

@workspace How do I add authentication to this project?

Inline Completion

As you type, Continue provides intelligent suggestions. Press Tab to accept.

Edit Code

Edit this function to handle errors properly

Pros

• ✅ Completely free and open-source
• ✅ Supports multiple IDEs
• ✅ Privacy-focused with local model support
• ✅ Flexible model selection
• ✅ Active community
• ✅ No vendor lock-in

Cons

• ❌ Less polished than commercial alternatives
• ❌ Requires configuration
• ❌ Smaller ecosystem
• ❌ May need technical setup for local models
• ❌ Fewer integrations

When to Use

• Privacy-conscious developers
• Teams needing model flexibility
• VS Code and JetBrains users
• Developers wanting open-source tools
• Projects with local model requirements

Use Cases

Use Case	Why Continue
Privacy-First Coding	Run entirely locally with Ollama/LM Studio
Multi-IDE Support	Works with VS Code and JetBrains IDEs
Model Flexibility	Switch between Claude, GPT, Gemini, local models
Open-Source Preference	Fully transparent, no vendor lock-in

Comparison with Alternatives

Feature	Continue	Roo Code	Copilot	Cursor
Cost	Free	Free	Paid	Freemium
Open Source	✅ Yes	✅ Yes	❌ No	❌ No
Local Models	✅ Yes	✅ Yes	❌ No	⚠️ Limited
JetBrains	✅ Yes	❌ No	✅ Yes	❌ No
VS Code	✅ Yes	✅ Yes	✅ Yes	❌ Fork
Privacy	✅ Excellent	✅ Excellent	⚠️ Cloud	⚠️ Cloud
Learning Curve	Low	Low	Low	Low
Best for	Multi-IDE, privacy	VS Code only	Enterprise	Daily coding

Best Practices

1. Configure models first — Set up preferred AI provider in config.json
2. Use local models for privacy — Run with Ollama for complete data control
3. Leverage @workspace context — Use @workspace for project-wide questions
4. Customize tab autocomplete — Configure dedicated model for completions
5. Review before accepting — Always verify AI-generated code before accepting

Troubleshooting

Issue	Solution
Completion not working	Check provider API key and model configuration
Chat not responding	Verify model is available and API is accessible
JetBrains plugin fails	Ensure extension is enabled in plugin settings
Local model fails	Verify Ollama/LM Studio is running with correct model

Resources

• Official Website
• GitHub
• Documentation