Update README.md

README.md

@@ -14,10 +14,71 @@ datasets:
 library_name: transformers
 ---
 
-![image](./image.webp)
-# Uploaded  model
+# Zirel-R1: Optimized for Fast and Essential Reasoning
 
-- **Developed by:** Daemontatox
-- **License:** apache-2.0
-- **Finetuned from model :** unsloth/deepseek-r1-distill-qwen-32b
+![Zirel Logo](./image.webp)
+
+## Model Overview
+
+**Zirel-R1** is an advanced **reasoning-optimized** model designed for **short, fast, and necessary reasoning**, avoiding long and unnecessary computation. It surpasses **Cogito-R1** and **PathfinderAI S1** in efficiency, making it ideal for applications requiring structured logical inference and quick decision-making.
+
+- **Developed by:** Daemontatox 
+- **Model Series:** Zirel  
+- **Base Model:** `unsloth/deepseek-r1-distill-qwen-32b`  
+- **License:** Apache-2.0  
+- **Languages:** English  
+- **Finetuned on:** `Daemontatox/math_conv`  
+- **Library:** Transformers  
+
+## Key Features
+
+✅ **Fast and Concise Reasoning** – Delivers precise answers with minimal computational overhead.  
+✅ **Optimized for Short-Form Problem Solving** – Excels in extracting core insights efficiently.  
+✅ **Enhanced Logical Inference** – Ideal for applications in structured decision-making, math reasoning, and controlled AI workflows.  
+
+## Usage
+
+You can load the model using `transformers`:
+
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+model_name = "Daemontatox/Zirel-R1"
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForCausalLM.from_pretrained(model_name, device_map="auto")
+
+prompt = "What is the next number in the sequence: 2, 4, 8, 16?"
+inputs = tokenizer(prompt, return_tensors="pt").to("cuda")
+output = model.generate(**inputs, max_new_tokens=50)
+print(tokenizer.decode(output[0], skip_special_tokens=True))
+
+Performance
+
+Speed: 🚀 Optimized for rapid inference and low-latency responses.
+
+Accuracy: 🎯 Fine-tuned on high-quality mathematical and reasoning datasets.
+
+Efficiency: ⚡ Processes only the necessary information for an answer.
+
+
+Citation
+
+If you use Zirel-R1, please cite:
+
+@misc{daemontatox2025zirel,
+  author = {Daemontatox},
+  title = {Zirel-R1: Optimized for Fast and Essential Reasoning},
+  year = {2025},
+  publisher = {Hugging Face},
+  url = {https://huggingface.co/Daemontatox/Zirel-R1}
+}
+
+License
+
+This model is released under the Apache-2.0 License.
+
+
+---
+
+This template ensures your model card looks professional and informative on Hugging Face. Let me know if you need modifications!