✨ Build an AI Chatbot with Google Gemini API in Python

Create a folder called gemini-chatbot. Open it in VS Code. Create a new file chatbot.py. Open a terminal (Ctrl + ` ) and verify Python:

You should see Python 3.8 or higher.

Google provides an official Python SDK called google-generativeai. Install it:

This is the only library you need — it handles all the API communication.

This is where Gemini beats OpenAI — completely free, no credit card required.

1. Go to aistudio.google.com
2. Sign in with your Google account
3. Click "Get API key" on the left sidebar
4. Click "Create API key" → "Create in new project"
5. Copy the key (starts with AIza...)

Open chatbot.py and add this code. Replace YOUR_API_KEY with your real key:

Run it: python chatbot.py. You should see a Gemini response. You just used a state-of-the-art AI model for free!

For a real chatbot we need memory. Gemini's start_chat() method makes this easy — it automatically tracks all messages in the conversation:

Try asking "What's 25 × 8?" then "Now multiply that by 3" — Gemini will remember the first answer.

Give your bot a persona using system_instruction:

Now your chatbot has personality. Change the instruction to make it act as a chef, philosopher, or anything you can imagine.

For a more engaging feel, stream responses word-by-word using stream=True:

Now your chatbot responds in real-time, just like ChatGPT. Much more polished feel.