Frequently Asked Questions

Have questions? We have answers. Here are some of the most common questions we get about Gemini CLI.

What's the difference between the Gemini API and the Gemini CLI?

The Gemini API is the backend service provided by Google that allows programs to access the AI models. The Gemini CLI is a user-friendly, official command-line tool that uses the API, letting you interact with the models directly from your terminal without writing any extra code.

How do I get a Gemini API Key? Is it free?

You can get a free API key from Google AI Studio. Simply sign in with your Google account and create a new key. We have a detailed, step-by-step guide on our blog that walks you through the entire process.

What are the free usage limits for the Gemini API?

Google offers generous free usage limits for the Gemini API. As of 2025, you can make up to 60 requests per minute and process millions of characters per month without any charge. This is more than enough for most individual developers and small projects.

Is it safe to use my API key in the Gemini CLI?

Yes, when used correctly. The official Gemini CLI stores your API key locally and securely. However, you should treat your API key like a password - never share it or commit it to public repositories. The CLI typically stores it in your system's secure credential storage.

Can I use Gemini CLI on Windows?

Yes, Gemini CLI is fully compatible with Windows. You can install it via npm, pip, or download the standalone executable depending on which implementation you choose. Check our installation guides for Windows-specific instructions.

How can I use Gemini to analyze my local files?

Gemini CLI allows you to pass file contents directly to the model using input redirection or specific file flags. For example: gemini --file=mycode.py "Explain this code" or cat mycode.py | gemini "Explain this code". This lets you analyze code, logs, or any text file.

What's the best Gemini model to use for coding?

For coding tasks, Gemini 1.5 Pro is currently the best option as it has a large context window (up to 1 million tokens) and strong coding capabilities. For simpler tasks, Gemini 1.5 Flash offers good performance with faster response times. You can specify the model with the --model flag.

Can I fine-tune a model with Gemini CLI?

While Gemini CLI itself doesn't provide fine-tuning capabilities, you can use Google's Vertex AI platform to fine-tune Gemini models for your specific use cases. The CLI can then be configured to use your custom-tuned model endpoints.