🧠 Local OCR Streamlit App

This project is a local OCR (Optical Character Recognition) app built with Streamlit, enabling users to extract text from images using LLMs.

🔍 Features

• 📁 Upload images or PDFs for OCR processing
• 🤖 Use Ollama for local LLM-based OCR
• ☁️ Use Sambanova cloud API for OCR
• 🧠 Supports multiple models from Ollama
• 🧹 Clear chat history and temp files easily
• 💬 Chat-like interface to view OCR results

---

⚙️ Requirements

• Python 3.8+
• Streamlit
• Ollama
• API Key from Sambanova

---

📦 Installation

1. Clone the repository:

git clone https://github.com/drisskhattabi6/Images_OCR_Locally.git
cd Images_OCR_Locally

2. Install dependencies:

pip install -r requirements.txt

3. Install Ollama & Pull a Vision Model:

Make sure Ollama is installed and running.

ollama pull llama3.2-vision

⚠️ Note: Ollama only supports images, not PDFs.

4. Set up your .env file:

Create a .env file and add your Sambanova API key:

API_KEY=your_sambanova_api_key

🔐 Required: You must have an API key from Sambanova to use their cloud models.

---

🚀 Run the App

streamlit run app.py

APP Screenshot:

---

📁 Folder Structure

local-ocr-app/
├── app.py                  # Main Streamlit app
├── ocr.py                  # OCR logic (local and remote)
├── .env                    # Contains API key  (create the File)
├── requirements.txt        # Python dependencies
└── tmp/                    # Temporary folder for uploaded files (auto-created)

---

❓ Troubleshooting

• ❌ No models found in Ollama? Run: ollama pull llama3.2-vision

• ❌ Missing API Key for Sambanova? Make sure .env file contains:

  API_KEY=your_key_here