Configuration¶

Estimated time to read: 3 minutes

Configure POML in VS Code settings (Ctrl+, or Cmd+,).

Example Complete Configuration¶

{
  "poml.languageModel.provider": "openai",
  "poml.languageModel.model": "gpt-4o",
  "poml.languageModel.apiKey": "sk-your-api-key-here",
  "poml.languageModel.apiUrl": "https://api.openai.com/v1/",
  "poml.languageModel.temperature": 0.7,
  "poml.languageModel.maxTokens": 1500,
  "poml.scrollPreviewWithEditor": true,
  "poml.markEditorSelection": true,
  "poml.trace": "off"
}

For multiple providers with different API keys, which is useful when overriding the default provider setting in POML files:

{
  "poml.languageModel.provider": "openai",
  "poml.languageModel.model": "gpt-4o",
  "poml.languageModel.apiKey": {
    "openai": "sk-your-openai-key",
    "anthropic": "sk-ant-your-anthropic-key",
    "google": "your-google-key"
  },
  "poml.languageModel.apiUrl": {
    "openai": "https://api.openai.com/v1/",
    "microsoft": "https://your-resource.openai.azure.com/"
  }
}

Language Model Configuration¶

The following settings mainly control the language model used for POML testing feature within VSCode.

Language Model Provider¶

{
  "poml.languageModel.provider": "openai"
}

Options: openai, microsoft, anthropic, google
Default: openai

Model Name¶

{
  "poml.languageModel.model": "gpt-4o"
}

Default: gpt-4o
For Azure OpenAI, use the deployment name. For other providers, use the model code name.

Temperature¶

{
  "poml.languageModel.temperature": 0.5
}

Default: 0.5
Range: 0.0 to 2.0
Controls randomness in responses. Lower values are more deterministic.

Max Tokens¶

{
  "poml.languageModel.maxTokens": 2000
}

Default: 0 (unlimited)
Maximum number of completion tokens to generate.

API Key¶

{
  "poml.languageModel.apiKey": "your-api-key-here"
}

Or use provider-specific keys:

{
  "poml.languageModel.apiKey": {
    "openai": "sk-your-openai-key",
    "anthropic": "sk-ant-your-anthropic-key",
    "google": "your-google-key",
    "microsoft": "your-azure-key"
  }
}

Required for most providers. Keep this secure and never commit to version control.

The API key can be: - A string for a single key used across all providers - An object with provider-specific keys, useful when switching between providers or when prompts override the provider at runtime

API URL¶

{
  "poml.languageModel.apiUrl": "https://api.openai.com/v1/"
}

Or use provider-specific URLs:

{
  "poml.languageModel.apiUrl": {
    "openai": "https://api.openai.com/v1/",
    "microsoft": "https://westeurope.api.cognitive.microsoft.com/",
    "anthropic": "https://api.anthropic.com/"
  }
}

Examples: - OpenAI: https://api.openai.com/v1/ - Azure OpenAI: https://westeurope.api.cognitive.microsoft.com/openai - Custom OpenAI-compatible: https://api.example.com/v2/

The API URL can be: - A string for a single URL used across all providers - An object with provider-specific URLs, useful when different providers require different endpoints

Warning

If you are using Azure OpenAI and encounter a Resource not found error, you may want to change the configuration from https://xxx.cognitiveservices.azure.com/ to https://xxx.cognitiveservices.azure.com/openai or vice versa. Refer to Vercel AI Azure Provider for more details.

API Version¶

{
  "poml.languageModel.apiVersion": "2024-02-15-preview"
}

Optional - Mainly used for OpenAI and Azure OpenAI services.

Provider-Specific Examples¶

Azure OpenAI¶

{
  "poml.languageModel.provider": "microsoft",
  "poml.languageModel.model": "my-gpt4-deployment",
  "poml.languageModel.apiKey": "your-azure-api-key",
  "poml.languageModel.apiUrl": "https://your-resource.openai.azure.com/",
  "poml.languageModel.apiVersion": "2024-02-15-preview"
}

Anthropic Claude¶

{
  "poml.languageModel.provider": "anthropic",
  "poml.languageModel.model": "claude-3-5-sonnet-20241022",
  "poml.languageModel.apiKey": "your-anthropic-api-key"
}

Google Gemini¶

{
  "poml.languageModel.provider": "google",
  "poml.languageModel.model": "gemini-1.5-pro",
  "poml.languageModel.apiKey": "your-google-api-key"
}

Preview & Editor Settings¶

These features need further testing. Please report bugs if you need this feature but it does not work as expected.

Scroll Synchronization¶

{
  "poml.scrollPreviewWithEditor": true,
  "poml.scrollEditorWithPreview": true
}

Default: true
Synchronize scrolling between editor and preview panes.

Editor Selection¶

{
  "poml.markEditorSelection": true,
  "poml.doubleClickToSwitchToEditor": true
}

Default: true
Highlight current editor selection in preview and enable double-click navigation.

Development Settings¶

Debugging¶

{
  "poml.trace": "verbose"
}

Options: off, verbose
Default: off

Enable detailed tracing for troubleshooting.

Telemetry¶

{
  "poml.telemetry.connection": ""
}

Default: "" (empty)
Development setting for telemetry connection string.