1. Python CLI Installer

python install.py

Interactive command-line installer. Detects Sublime Text automatically, walks you through configuration.

2. GUI Installer

python install_gui.py

Graphical installer with Tkinter. Pre-loads existing settings, configurable interface.

Settings are saved automatically. Reconfigure Anytime with Ctrl+Shift+P. Following are configuration options.

• endpoint: Your OpenAI-compatible API endpoint (v1 format) - Required

  • OpenAI: https://api.openai.com/v1/chat/completions
  • Local server: http://localhost:8000/v1/chat/completions
  • Other providers: Use their v1-compatible endpoint

• model: The model to use for completions - Required

  • Tested models so far are as follows:
    • gpt-oss-20b
    • Qwen/Qwen2.5-Coder-1.5B-Instruct

• api_key: Authentication key (optional)

  • Only needed if your endpoint requires it
  • For OpenAI: sk-...
  • Leave blank if not needed

• max_context_lines: Number of lines of context to send (default: 30)

  • Increase for more context, decrease for faster responses

• timeout_ms: Request timeout in milliseconds (default: 30000)

  • Increase if using slower endpoints

• trigger_language: Array of language scopes to enable the plugin

  • Examples: python, cpp, javascript, typescript, java, go, rust, etc.

• debug: Enable debug logging to console (default: false)

  • Set to true to see detailed logs in View >> Show Console
  • Plugin is silent by default

Setup wizard not appearing

• Restart Sublime Text: File -> Exit, then reopen
• Check Sublime Text console (View -> Show Console) for errors
• Manually run: Preferences > Package Settings > CodeContinue > Settings
• Or use Ctrl+Shift+P >> "CodeContinue: Configure"

Suggestions not appearing

• Check that your language is in the trigger_language list
• Verify your API endpoint is accessible and correct
• Enable debug logging: set "debug": true in settings, then check View >> Show Console
• Try Ctrl+Shift+P -> "CodeContinue: Configure" to verify settings
• Make sure you have an active API key if required

Timeout errors

• Increase timeout_ms in settings (default: 20000ms)
• Try a faster model or local endpoint
• Check your internet connection
• Verify your API key is valid

Authentication/Connection errors

• Verify your API key is correct
• Make sure endpoint URL is exactly right (copy-paste to avoid typos)
• Check if the endpoint is currently running/available
• Use Ctrl+Shift+P -> "CodeContinue: Configure" to update credentials

Keybindings not working

• Make sure you've set up keybindings (they're not enabled by default)
• Go to Preferences > Package Settings > CodeContinue > Key Bindings
• Check for conflicts with other packages
• Try alternative key combinations

OpenAI:

endpoint: https://api.openai.com/v1/chat/completions
model: gpt-3.5-turbo or gpt-4
api_key: sk-...

Local LLM (LLaMA, Mistral, etc.):

endpoint: http://localhost:8000/v1/chat/completions
model: (whatever model you're running)
api_key: (usually not needed)

Hugging Face Inference API:

endpoint: https://api-inference.huggingface.co/v1/chat/completions
model: HuggingFaceH4/zephyr-7b-beta
api_key: hf_...

Other providers: Any endpoint supporting OpenAI's v1 chat completion format will work.