🤖 Gemini Chatbot

A sleek, conversational AI chatbot powered by Google Gemini 2.5 Flash and built with Streamlit — featuring full multi-turn memory so it actually remembers what you said.

---

✨ Features

• 💬 Multi-turn conversations — full chat history passed on every request
• ⚡ Gemini 2.5 Flash — fast, capable, and cutting-edge
• 🧠 Session memory — context is preserved across the entire conversation
• 🖥️ Clean Streamlit UI — zero-fuss chat interface, ready out of the box
• 🔐 Secure API key handling — via .env and python-dotenv

---

🚀 Getting Started

1. Clone the repo

git clone https://github.com/your-username/gemini-chatbot.git
cd gemini-chatbot

2. Install dependencies

pip install streamlit google-genai python-dotenv

3. Set up your API key

Create a .env file in the project root:

GEMINI_API=your_google_gemini_api_key_here

Get your key at Google AI Studio

4. Run the app

streamlit run app.py

---

📁 Project Structure

gemini-chatbot/
├── app.py          # Streamlit web chatbot (main UI app)
├── main.py         # Terminal chatbot (practice CLI version, no Streamlit)
├── .gitignore
└── README.md

---

🔧 How It Works

Each time the user sends a message, the full conversation history is formatted and passed to the Gemini API — giving the model complete context for a coherent, memory-aware response.

response = client.models.generate_content(
    model="gemini-2.5-flash",
    contents=[f"{m['role']}: {m['content']}" for m in st.session_state.messages]
)

---

⚠️ Important

Never commit your .env file. Add it to .gitignore:

.env

---

🛠️ Built With

Tool	Purpose
Streamlit	UI framework
Google Gemini	LLM backend
python-dotenv	Env management

---

👤 Author

Amitava Biswas

• GitHub: @AmitavaBiswas

---