https://github.com/yahya-py/ollama-llm-rag-chatbot
Local AI chatbot using LLaMA3 + personal data with LangChain, Chroma & Gradio UI
https://github.com/yahya-py/ollama-llm-rag-chatbot
chatbot gradio langchain llama3 local localllm python
Last synced: 3 months ago
JSON representation
Local AI chatbot using LLaMA3 + personal data with LangChain, Chroma & Gradio UI
- Host: GitHub
- URL: https://github.com/yahya-py/ollama-llm-rag-chatbot
- Owner: Yahya-py
- Created: 2025-07-30T19:14:47.000Z (12 months ago)
- Default Branch: main
- Last Pushed: 2025-07-31T09:14:08.000Z (11 months ago)
- Last Synced: 2025-09-27T13:29:20.683Z (10 months ago)
- Topics: chatbot, gradio, langchain, llama3, local, localllm, python
- Language: Python
- Homepage: https://github.com/Yahya-py/OLLAMA-LLM-RAG-Chatbot
- Size: 1.12 MB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# 🤖 Smart AI Chatbot with Local LLM + Personal Data
This project is a lightweight, fully functional AI chatbot built with **Gradio**, **LangChain**, and **Ollama**. It supports both general-purpose chatting using local LLMs (like `llama3.2`) and personalized answers based on your own uploaded documents — all without relying on cloud APIs like OpenAI or Cohere.
Smart AI Chatbot built on LLaMA3 using LangChain, ChromaDB, and Gradio. Supports real-time chat with optional retrieval-augmented responses from your own documents. 100% offline and privacy-first — perfect for developers, researchers, and businesses
---
## 🔍 Features
- 🧠 **Local LLM** (e.g. `llama3.2`) powered via [Ollama](https://ollama.com/)
- 📚 **Document-aware Q&A** using `langchain` + `Chroma` vector DB
- 💬 **Gradio Interface** with full chat history
- 🖼️ Custom branding with logo and responsive UI
- ✅ Toggle between general chatbot and personal knowledge base
- 🔒 Runs 100% offline — fully private and secure
---
## 📸 Preview
---
## 🚀 How It Works
| Mode | Description |
|------|-------------|
| **General Mode** | Chatbot answers general questions using your chosen local LLM |
| **Personal Data Mode** | It retrieves relevant info from your uploaded `.txt` files before generating a response |
---
## 🛠️ Installation
### 1. Clone this repository
```bash
git clone https://github.com/Yahya-py/OLLAMA-LLM-RAG-Chatbot.git
cd OLLAMA-LLM-RAG-Chatbot
```
### 2. Install requirements
```bash
pip install -r requirements.txt
```
### 3. Install and Run Ollama
```bash
ollama pull llama3.2
ollama run llama3.2
```
---
## 📁 Directory Structure
```
OLLAMA-LLM-RAG-Chatbot/
├── chatbot_ollama_rag_llama32.py # Main Python script
├── image.png # Logo for chatbot UI
├── data/ # Folder containing user documents (.txt)
├── chroma_db/ # Persisted vector DB (auto-generated)
└── requirements.txt
```
---
## ⚙️ Usage
```bash
python chatbot_ollama_rag_llama32.py
```
Visit [http://127.0.0.1:7860](http://127.0.0.1:7860) in your browser.
You can:
- Type your question
- Check "Use Personal Documents?" to answer from your documents
- View full chat history in real-time
---
## 🧠 Customize
- Replace `image.png` with your own brand logo
- Add `.pdf`, `.docx`, or `.md` support using `langchain` loaders
- Train or fine-tune your own model using [Ollama fine-tune](https://ollama.com/blog/fine-tune-your-own-model)
---
## 🔐 Privacy Note
This project **does not send any data to the cloud**. All inference and document processing are done locally.
---
## 🏷️ Tags
```
python, chatbot, langchain, gradio, ollama, llama3, chromadb, local-llm, document-chatbot, AI-portfolio
```
---
## 📃 License
MIT License — free to use, modify, and share.