Update README.md

README.md

@@ -1,10 +1,44 @@
 ---
-title: Platform
-emoji: 🌍
-colorFrom: pink
-colorTo: pink
+title: Open CaptchaWorld
+emoji: 🧩
+colorFrom: blue
+colorTo: green
 sdk: docker
+sdk_version: 3.9
+app_file: app.py
 pinned: false
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# Open CaptchaWorld Benchmark
+
+Open CaptchaWorld is a comprehensive benchmark for evaluating intelligence through diverse visual reasoning challenges. It features 20 different types of CAPTCHA puzzles, each testing different aspects of visual reasoning.
+
+## Overview
+
+This benchmark contains a variety of CAPTCHA types including:
+
+- **Dice Count**: Count the sum of numbers on dice
+- **Geometry Click**: Identify and click on geometric shapes 
+- **Rotation Match**: Rotate objects to match reference positions
+- **Unusual Detection**: Find unusual elements in a grid
+- **Image Recognition**: Select images matching a description
+- **And many more!**
+
+Each puzzle type is designed to test different reasoning abilities and challenges both humans and AI systems.
+
+## How to Use
+
+1. The benchmark will present you with a series of CAPTCHA puzzles
+2. Read the instructions for each puzzle carefully
+3. Submit your answer using the provided interface
+4. View your accuracy statistics at the top of the page
+
+## Difficulty Levels
+
+Each puzzle type has a "Reasoning Depth" rating from 1-5 stars, indicating its complexity level.
+
+## About
+
+Open CaptchaWorld was created to advance research in visual reasoning and intelligence measurement. It provides a standardized way to evaluate both human and AI capabilities across a diverse range of visual reasoning tasks.
+
+[View the full GitHub repository](https://github.com/louvain-media/Open_CaptchaWorld)