Update README.md

README.md

@@ -23,4 +23,127 @@ tags:
 - qwen
 - educational
 - fine-tuned
----
+---
+
+<style>
+body {
+  font-family: 'Segoe UI', sans-serif;
+  line-height: 1.7;
+  color: #1e1e1e;
+  background: #ffffff;
+  padding: 1.5em;
+}
+
+h1, h2, h3 {
+  color: #2c3e50;
+  border-bottom: 2px solid #eee;
+  padding-bottom: 5px;
+}
+
+img {
+  display: block;
+  margin: 1em auto;
+  max-width: 200px;
+}
+
+code, pre {
+  background-color: #f5f5f5;
+  padding: 0.5em;
+  border-radius: 5px;
+  font-family: 'Courier New', monospace;
+  display: block;
+  white-space: pre-wrap;
+}
+
+blockquote {
+  border-left: 4px solid #3498db;
+  background: #ecf6fd;
+  padding: 0.8em;
+  color: #333;
+  margin: 1.5em 0;
+}
+</style>
+
+# 🤖 إدراكي (Edraky) - Multilingual Educational AI Model
+
+![Edraky Logo](https://cdn-uploads.huggingface.co/production/uploads/686e726239f003427404a1be/uuB7LFKDX1C5B28DGJyZN.png)
+
+> A fine-tuned AI assistant that helps Egyptian students — specially 3rd preparatory — with Arabic, English, Hebrew content, and educational support.
+
+---
+
+## 🧠 Model Details
+
+- **Base Model:** Qwen/Qwen2-1.5B-Instruct
+- **Languages:** Arabic, English, Hebrew
+- **Trained on:** Educational, Q&A, mathematical, and multilingual prompts
+- **License:** Apache-2.0
+- **Tags:** multilingual, Arabic, Hebrew, educational, fine-tuned, transformers
+
+---
+
+## 🚀 How to Use
+
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+model = AutoModelForCausalLM.from_pretrained("Edraky/Edraky")
+tokenizer = AutoTokenizer.from_pretrained("Edraky/Edraky")
+
+input_text = "اشرح الثورة العرابية"
+inputs = tokenizer(input_text, return_tensors="pt")
+outputs = model.generate(**inputs, max_new_tokens=100)
+print(tokenizer.decode(outputs[0]))
+✅ Intended Uses
+🧑‍🏫 Educational chatbot for classroom topics
+
+📚 Answering curriculum-based questions
+
+✍️ Writing, completion, and explanation for Arabic texts
+
+🏫 Perfect fit for 3rd preparatory grade in Egypt
+
+🚫 Limitations / Out-of-Scope
+❌ Not designed for real-time tutoring
+
+❌ No legal, political, or medical advice
+
+❌ Not for biased, violent, or harmful use
+
+📊 Evaluation & Training
+Datasets:
+
+fka/awesome-chatgpt-prompts
+
+microsoft/rStar-Coder
+
+gsm8k-rerun/Qwen_Qwen2.5-1.5B-Instruct
+
+Metrics Used:
+
+Accuracy
+
+Perplexity
+
+Word Error Rate (WER)
+
+🌍 Environment Impact (Optional)
+Trained on GPU (details coming soon). Expected carbon footprint is low due to small model size (1.5B).
+
+👨‍💻 Maintainers & Contact
+Created by: Edraky Team
+
+Contact: edraky.edu@gmail.com
+
+License: Apache 2.0
+
+📜 Citation (if needed)
+bibtex
+Copy
+Edit
+@misc{edraky2025,
+  title={Edraky: Multilingual Educational AI Model},
+  author={Edraky Team},
+  year={2025},
+  howpublished={\\url{https://huggingface.co/Edraky/Edraky}}
+}