---
title: Demo DeepSeek OCR
emoji: 🦀
colorFrom: indigo
colorTo: red
sdk: gradio
sdk_version: 5.49.1
app_file: app.py
pinned: false
license: mit
short_description: Aplicación OCR inteligente basada en DeepSeek-AI,AB
---

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

# 🧠 Demo DeepSeek OCR (Light Version)

Welcome to **Demo DeepSeek OCR**, an experimental project developed by **Andrey Beltrán** to explore the power of **open-source AI models** for Optical Character Recognition (OCR).

This lightweight demo uses **EasyOCR** for text extraction, providing a simple and efficient alternative when GPU resources are not available.  
It serves as a **proof of concept** for building and deploying vision-based AI applications on **Hugging Face Spaces** using only CPU.

---

## 🚀 Features

- 📝 Extract text from images with high precision  
- 🧾 Convert image content into Markdown format  
- 🟥 Draw bounding boxes for computer vision tasks  
- 🧠 Summarize or explain image content (ideal for documents, charts, and tables)

---

## ⚙️ Technology Stack

| Component | Description |
|------------|-------------|
| **Framework** | [Gradio](https://gradio.app) – Interactive UI for machine learning demos |
| **OCR Engine** | [EasyOCR](https://github.com/JaidedAI/EasyOCR) – Lightweight open-source OCR library |
| **Hosting** | [Hugging Face Spaces](https://huggingface.co/spaces) – Free platform for ML demos |
| **Language Support** | English 🇺🇸 and Spanish 🇪🇸 |
| **Hardware** | CPU-only (no GPU required) |

---

## 🧩 Project Structure