CLI AI Chatbot using Ollama

A local AI-powered CLI chatbot built using Python and Ollama with persistent memory support.

This chatbot:

• runs completely offline
• stores conversation history
• supports multi-turn conversations
• uses local LLMs through Ollama
• requires no API key

---

Features

Local AI Chatbot

Runs AI models locally using Ollama.

Persistent Memory

Conversation history is stored in memory.json so the chatbot remembers previous chats.

Multi-turn Conversations

The chatbot maintains context across messages.

Offline Usage

No internet or API key required after model download.

Lightweight CLI Interface

Simple terminal-based chatbot.

---

Tech Stack

Technology	Purpose
Python	Programming language
Ollama	Local LLM runtime
JSON	Persistent memory storage

---

Project Structure

CLI_Chatbot/
│
├── main.py
├── memory.json
├── requirements.txt
├── README.md
└── .gitignore

---

Requirements

Install Python

Download Python: https://www.python.org/downloads/

---

Install Ollama

Download Ollama: https://ollama.com/download

---

Install Dependencies

pip install -r requirements.txt

---

Download a Model

Example:

ollama run llama3.2:3b

This downloads the model locally.

You can also use:

• llama3
• mistral
• gemma:2b

---

Run the Chatbot

python main.py

---

Example Usage

Chat Started...
Type 'exit' to end the conversation.

You: Hello
Assistant: Hi! How can I help you today?

You: My name is Aditya
Assistant: Nice to meet you, Aditya!

You: What is my name?
Assistant: Your name is Aditya.

---

How Memory Works

Conversation history is stored inside:

memory.json

Every message:

1. gets added to conversation history
2. is sent to the AI model
3. gets saved permanently

This allows the chatbot to remember previous conversations even after restarting.

---

Main Concepts Used

System Role

Defines chatbot behavior.

Example:

{
    "role": "system",
    "content": "You are a helpful assistant."
}

---

User Role

Stores user messages.

Example:

{
    "role": "user",
    "content": "Hello"
}

---

Assistant Role

Stores chatbot replies.

Example:

{
    "role": "assistant",
    "content": "Hi!"
}

---

Future Improvements

Possible upgrades:

• Streaming responses
• Semantic memory
• SQLite database
• Voice assistant
• Rich terminal UI
• Tool usage
• GUI interface
• Vector databases

---

Resume Value

This project demonstrates:

• Python development
• Conversational AI systems
• Local LLM integration
• Persistent memory handling
• CLI application development
• Context management

---

.gitignore

Recommended .gitignore:

venv/
__pycache__/
*.pyc
memory.json

---

License

This project is open-source and intended for educational purposes.

---

Author

Built by Aditya Jhinjha