Local-AI

Automate the process of using Ollama AI models with a local HTML / CSS / JavaScript web interface and a Python backend. The system detects hardware, selects the best model, installs it if missing, and starts a local web server.

---

Goal of the project

• Automate Ollama AI model usage
• Select the best model based on hardware
• Provide a local web interface
• Support Windows and Linux
• One-command startup

---

How It Works

• User runs the startup script
• Prerequisites are checked
• Hardware information is collected
• Best model is selected using model_config.json
• User may override the model
• Ollama installs the model if missing
• Python web server starts
• Web UI becomes available locally (localhost / LAN)
• Logs server errors and shows output in server_errors.log

---

CLI Inteface

---

Web Interface (Pc)

---

Execution

Windows:
start.bat

Linux:
bash start.sh

Project Structure

Local-AI/
├── start.sh
├── start.bat
├── server.py
├── model_config.json
│── index.html
│── style.css
└── script.js

To-do

• Catch and fix , Error: listen tcp 127.0.0.1:11434: bind: address already in use
• Make the UI , phone friendly
• Add more AI models in model_config.json file