Skip to content
CentralMind

Google Gemini

The Gemini provider enables direct integration with Google’s language models through a compatible OpenAI-style API. This guide covers setup, configuration, authentication, and usage options for connecting to the Gemini API.

Overview

The Gemini provider allows access to Google’s language models via an OpenAI-compatible interface. This provider offers streamlined integration with authentication through API keys, flexible configuration options, and support for advanced features.

Authentication

The provider uses API key authentication to access Gemini’s services. You’ll need to obtain a Gemini API key to use this provider:

  • Environment Variable: Set GEMINI_API_KEY to provide your credentials.
  • Command Line: Pass your API key using the appropriate flag.

Getting an API Key

Google offers a free tier for Gemini API access. You can obtain an API key by visiting Google AI Studio:

Once logged in, you can create an API key in the API section of AI Studio. The free tier includes a generous monthly token allocation, making it accessible for development and testing purposes.

Example Usage

Terminal window
export GEMINI_API_KEY='yourkey'

Below is a basic example of how to use the provider with a connection configuration file:

Terminal window
./gateway discover \
  --ai-provider gemini \
  --config connection.yaml

Model Selection

By default, the Gemini provider uses gemini-2.0-flash-thinking-exp. You can specify a different model using one of the following methods:

  1. Command-line Flag: Use the --ai-model flag.
  2. Environment Variable: Set the GEMINI_MODEL_ID.

Examples:

Terminal window
# Specify model via command line
./gateway discover \
  --ai-provider gemini \
  --ai-model gemini-2.0-flash-thinking-exp \
  --config connection.yaml


# Or via environment variable
export GEMINI_MODEL_ID=gemini-2.0-flash-thinking-exp
./gateway discover \
  --ai-provider gemini \
  --config connection.yaml

Advanced Configuration

Response Length Control

Control the maximum token count in responses:

Terminal window
./gateway discover \
  --ai-provider gemini \
  --ai-max-tokens 8192 \
  --config connection.yaml

If not specified, the default maximum token count is 100,000 tokens.

Temperature Adjustment

Adjust the randomness of responses with the temperature parameter:

Terminal window
./gateway discover \
  --ai-provider gemini \
  --ai-temperature 0.7 \
  --config connection.yaml

Lower values produce more deterministic outputs, while higher values increase creativity and randomness.

Usage Costs

Google offers a free tier for Gemini API usage with monthly token limits. For production workloads or higher usage requirements, please refer to Google’s documentation for the current pricing for Gemini models.

  • API Key Security: Use environment variables for sensitive settings like API keys.
  • Manage Token Count: Set a reasonable maximum token count to control costs.
  • Select the Right Model: Choose models based on your specific needs and budget constraints.
  • Start with Free Tier: Use the free tier to experiment and prototype before committing to paid usage.

Additional Resources

For more information about Google’s language models and their capabilities, please refer to Google’s official documentation and AI Studio resources.