🤖 Local LLM Chat App (Ollama + Llama 3 + LangChain)

A simple, fast, and clean AI chatbot built using:

Ollama (local LLM runner)

Llama 3 model

LangChain LCEL

Streamlit UI

This project shows how to run a real offline LLM chatbot with a modern, chat-style interface.

🚀 Features

🧠 Runs Llama 3 locally using Ollama

💬 Clean chat UI using Streamlit

🔗 Modern LangChain LCEL pipeline (prompt | llm | parser)

💾 Session-based chat history

🎚 Sidebar with model selection

⚡ Instant, smooth responses

🔒 Safe environment variable handling using .env

📂 Project Structure local-llm-chat-app/ │ ├── app.py ├── README.md ├── requirements.txt ├── .env.example ├── .gitignore

🛠 Installation 1️⃣ Create virtual environment (optional but recommended) python -m venv venv source venv/bin/activate # Mac/Linux venv\Scripts\activate # Windows

2️⃣ Install dependencies pip install -r requirements.txt

🔧 Environment Variables

Duplicate .env.example and rename it to:

.env

Add your keys:

LANGCHAIN_API_KEY=your_langchain_api_key_here LANGCHAIN_PROJECT=your_project_name_here

(These are optional — the app works even without LangSmith.)

🦙 Install & Run Ollama

Install Ollama from: https://ollama.com

Pull the llama3 model:

ollama pull llama3

Or use:

ollama pull gemma:2b

▶️ Running the App

Start Streamlit:

streamlit run app.py

Your LLM chatbot will open in the browser 🎉

🧠 Technologies Used

Python

LangChain

LangChain Ollama integration

Streamlit

Llama 3 (via Ollama)

dotenv (environment management)

👨‍💻 Author

Shehjad Patel AI Engineer | LangChain | LLMs | Streamlit | Ollama