Azure OpenAI¶

Configure HolmesGPT to use Azure OpenAI Service.

Setup¶

Configuration¶

Required Environment Variables¶

export AZURE_API_VERSION="2024-02-15-preview"
export AZURE_API_BASE="https://my-org.openai.azure.com/"
export AZURE_API_KEY="your-azure-api-key"  # Optional, can use --api-key instead

Usage¶

holmes ask "what pods are unhealthy and why?" --model=azure/<DEPLOYMENT_NAME> --api-key=<API_KEY>

Replace <DEPLOYMENT_NAME> with your actual Azure OpenAI deployment name.

Environment Variables¶

AZURE_API_VERSION - API version (e.g., 2024-02-15-preview)
AZURE_API_BASE - Your Azure endpoint URL (e.g., https://my-org.openai.azure.com/)
AZURE_API_KEY - Your Azure API key (optional if passed via --api-key)

Available Models¶

Azure OpenAI uses deployment names instead of model names:

# Using gpt-4 deployment
holmes ask "analyze cluster issues" --model=azure/gpt-4-deployment

# Using gpt-35-turbo deployment
holmes ask "quick cluster check" --model=azure/gpt-35-turbo-deployment

Note

Model availability varies by region. Check your Azure portal for available models.

Troubleshooting¶

Authentication Errors

Error: Azure authentication failed

- Verify environment variables are set correctly - Check that your Azure OpenAI resource is deployed - Ensure your API key has the correct permissions

Deployment Not Found

Error: Deployment not found

- Verify your deployment name is correct - Check that the deployment is in the same region as your API base - Ensure the deployment is fully deployed and running

Regional Issues

Error: Model not available in region

- Some models are only available in specific regions - Consider deploying in a different region