​

Continue.dev

​

Overview

• 200+ models: Access OpenAI, Anthropic, Google, Cohere, and more
• Cost savings: Leverage Eden AI’s competitive pricing
• Provider flexibility: Switch models instantly without changing configuration

​

Installation

​

VS Code

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

​

JetBrains IDEs

1. Open your JetBrains IDE (IntelliJ, PyCharm, etc.)
2. Go to Plugins (Ctrl+Alt+S / Cmd+,)
3. Search for “Continue”
4. Click Install

​

Configuration

​

Step 1: Open Configuration

1. Click the Continue icon in the sidebar
2. Click the gear icon (⚙️) to open settings
3. This opens ~/.continue/config.json

​

Step 2: Configure Eden AI

​

Step 3: Save and Reload

1. Save the configuration file
2. Reload your IDE or click “Reload” in Continue

​

Available Models

​

Chat Models

​

Features

​

Chat

1. Select your model from the dropdown
2. Type your question or request
3. Get instant responses

• “Explain this code”
• “Add error handling to this function”
• “Write tests for this component”

​

Inline Editing

1. Highlight code
2. Press Cmd+I (Mac) / Ctrl+I (Windows/Linux)
3. Describe your changes
4. Review and accept

• “Add type hints”
• “Refactor this to use async/await”
• “Simplify this logic”

​

Tab Autocomplete

1. Start typing
2. Continue suggests completions
3. Press Tab to accept

​

Slash Commands

• /edit - Edit selected code
• /comment - Add comments
• /test - Generate tests
• /fix - Fix errors
• /explain - Explain code

​

Advanced Configuration

​

Custom Context Providers

​

Model-Specific Settings

​

Environment Variables

​

Use Cases

​

Code Generation

1. Open Continue chat
2. Describe what you need: “Create a Python class for user authentication with login, logout, and token refresh methods”
3. Review and insert the generated code

​

Code Review

1. Select code
2. Open Continue chat
3. Ask: “Review this code for security issues and performance improvements”

​

Documentation

1. Select a function
2. Use /comment command
3. Continue adds comprehensive docstrings

​

Debugging

1. Copy error message
2. Open Continue chat
3. Ask: “How do I fix this error?” and paste the error

​

Refactoring

1. Select old code
2. Use /edit command
3. Ask: “Refactor this to use modern Python features”

​

Troubleshooting

​

Models Not Appearing

1. Check config.json syntax is valid (use a JSON validator)
2. Ensure apiBase is exactly: https://api.edenai.run/v3/llm
3. Reload the IDE

​

Authentication Errors

1. Verify your Eden AI API key in config.json
2. Check there are no extra spaces in the API key
3. Test the API key manually:

curl -X POST https://api.edenai.run/v3/llm/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "anthropic/claude-3-5-sonnet-20241022",
    "messages": [{"role": "user", "content": "test"}],
    "stream": true
  }'

​

Slow Responses

1. Use faster models for autocomplete (gpt-3.5-turbo)
2. Reduce maxTokens in completion options
3. Check your internet connection

​

Context Issues

1. Add relevant context providers
2. Include specific files using @file in chat
3. Use folder context for project-wide understanding

​

Keyboard Shortcuts

​

VS Code

• Cmd/Ctrl + I - Inline edit
• Cmd/Ctrl + Shift + L - Open Continue sidebar
• Cmd/Ctrl + Shift + R - Re-run last command

​

JetBrains

• Alt + Enter - Show Continue actions
• Cmd/Ctrl + Shift + C - Open Continue chat

​

Best Practices

​

1. Choose the Right Model

• Quick edits: Use openai/gpt-3.5-turbo for speed
• Complex reasoning: Use anthropic/claude-3-5-sonnet-20241022
• Autocomplete: Use openai/gpt-3.5-turbo for low latency

​

2. Provide Context

• Tag files with @filename
• Include error messages
• Describe the broader goal

​

3. Iterate

• Review generated code
• Ask for adjustments
• Combine AI suggestions with your expertise

​

4. Use Slash Commands

• /edit for modifications
• /test for test generation
• /fix for error resolution

​

Example Workflows

​

Building a New Feature

1. Chat: “Help me design a REST API endpoint for user registration”
2. Review the suggested structure
3. Use /edit to implement the endpoint
4. Use /test to generate unit tests
5. Use /comment to add documentation

​

Debugging

1. Copy the error traceback
2. Chat: “Explain this error and suggest a fix: [paste error]”
3. Review the explanation
4. Use /fix on the problematic code
5. Test the solution

​

Code Review

1. Select a function
2. Chat: “Review this for security, performance, and maintainability”
3. Review suggestions
4. Use /edit to apply improvements
5. Use /test to ensure nothing broke

​

Next Steps

• VS Code Integration - More VS Code AI tools
• Claude Code - Official Claude CLI
• Cline - Another powerful VS Code extension
• API Reference - Eden AI API documentation