Gemini

Configure HolmesGPT to use Google's Gemini models via Google AI Studio.

Setup

Get your API key from Google AI Studio.

Configuration

export GEMINI_API_KEY="your-gemini-api-key"
holmes ask "what pods are failing?" --model="gemini/<your-gemini-model>"

Using CLI Parameters

You can also pass the API key directly as a command-line parameter:

holmes ask "what pods are failing?" --model="gemini/<your-gemini-model>" --api-key="your-api-key"

Additional Resources

HolmesGPT uses the LiteLLM API to support Gemini provider. Refer to LiteLLM Gemini docs for more details.