Update Doc.

Author: codewithdark-git

README.md

@@ -1,44 +1,33 @@
 <div align="center">
-  <img src="docs/images/logo.png" alt="QuantLLM Logo" />
+  <img src="docs/images/logo.png" alt="QuantLLM Logo" width="150"/>
 
   # 🚀 QuantLLM v2.0
 
-  <p align="center">
-    <strong>The Ultra-Fast LLM Quantization & Export Library</strong>
-  </p>
-
-  <p align="center">
-    <a href="https://pepy.tech/projects/quantllm"><img src="https://static.pepy.tech/badge/quantllm" alt="Downloads"/></a>
-    <img alt="PyPI - Version" src="https://img.shields.io/pypi/v/quantllm?logo=pypi&label=version&color=orange"/>
-    <img alt="Python" src="https://img.shields.io/badge/python-3.10+-orange.svg"/>
-    <img alt="License" src="https://img.shields.io/badge/license-MIT-orange.svg"/>
-    <img alt="Stars" src="https://img.shields.io/github/stars/codewithdark-git/QuantLLM?style=social"/>
-  </p>
-
-  <p align="center">
-    <b>Load → Quantize → Fine-tune → Export</b> — All in One Line
-  </p>
+  **The Ultra-Fast LLM Quantization & Export Library**
+
+  [![Downloads](https://static.pepy.tech/badge/quantllm)](https://pepy.tech/projects/quantllm)
+  [![PyPI](https://img.shields.io/pypi/v/quantllm?logo=pypi&label=version&color=orange)](https://pypi.org/project/quantllm/)
+  [![Python](https://img.shields.io/badge/python-3.10+-orange.svg)](https://www.python.org/)
+  [![License](https://img.shields.io/badge/license-MIT-orange.svg)](LICENSE)
+  [![Stars](https://img.shields.io/github/stars/codewithdark-git/QuantLLM?style=social)](https://github.com/codewithdark-git/QuantLLM)
+
+  **Load → Quantize → Fine-tune → Export** — All in One Line
 
-  <p align="center">
-    <a href="#-quick-start">Quick Start</a> •
-    <a href="#-features">Features</a> •
-    <a href="#-export-formats">Export Formats</a> •
-    <a href="#-examples">Examples</a> •
-    <a href="https://quantllm.readthedocs.io">Documentation</a>
-  </p>
+  [Quick Start](#-quick-start) • 
+  [Features](#-features) • 
+  [Export Formats](#-export-formats) • 
+  [Examples](#-examples) • 
+  [Documentation](https://quantllm.readthedocs.io)
+
 </div>
 
 ---
 
 ## 🎯 Why QuantLLM?
 
-<table>
-<tr>
-<td width="50%">
+### ❌ Without QuantLLM (50+ lines of code)
 
-### ❌ Without QuantLLM
 ```python
-# 50+ lines of configuration...
 from transformers import AutoModelForCausalLM, BitsAndBytesConfig
 from peft import LoraConfig, get_peft_model
 import torch
@@ -53,44 +42,29 @@ model = AutoModelForCausalLM.from_pretrained(
     "meta-llama/Llama-3-8B",
     quantization_config=bnb_config,
     device_map="auto",
-    # ... more config
 )
 # Then llama.cpp compilation for GGUF...
 # Then manual tensor conversion...
 ```
 
-</td>
-<td width="50%">
+### ✅ With QuantLLM (4 lines of code)
 
-### ✅ With QuantLLM
 ```python
 from quantllm import turbo
 
-# One line does everything
-model = turbo("meta-llama/Llama-3-8B")
-
-# Generate
-print(model.generate("Hello!"))
-
-# Fine-tune
-model.finetune(dataset, epochs=3)
-
-# Export to any format
-model.export("gguf", quantization="Q4_K_M")
+model = turbo("meta-llama/Llama-3-8B")     # Auto-quantizes
+model.generate("Hello!")                    # Generate text
+model.export("gguf", quantization="Q4_K_M") # Export to GGUF
 ```
 
-</td>
-</tr>
-</table>
-
 ---
 
 ## ⚡ Quick Start
 
 ### Installation
 
 ```bash
-# Recommended installation
+# Recommended
 pip install git+https://github.com/codewithdark-git/QuantLLM.git
 
 # With all export formats
@@ -102,14 +76,14 @@ pip install "quantllm[full] @ git+https://github.com/codewithdark-git/QuantLLM.g
 ```python
 from quantllm import turbo
 
-# Load any model with automatic optimization
+# Load with automatic optimization
 model = turbo("meta-llama/Llama-3.2-3B")
 
 # Generate text
 response = model.generate("Explain quantum computing simply")
 print(response)
 
-# Export to GGUF for Ollama/llama.cpp
+# Export to GGUF
 model.export("gguf", "model.Q4_K_M.gguf", quantization="Q4_K_M")
 ```
 
@@ -123,79 +97,59 @@ model.export("gguf", "model.Q4_K_M.gguf", quantization="Q4_K_M")
 
 ## ✨ Features
 
-<table>
-<tr>
-<td width="50%">
-
 ### 🔥 TurboModel API
+
+One unified interface for everything:
+
 ```python
-# One unified API for everything
 model = turbo("mistralai/Mistral-7B")
 model.generate("Hello!")
 model.finetune(data, epochs=3)
 model.export("gguf", quantization="Q4_K_M")
 model.push("user/repo", format="gguf")
 ```
 
-</td>
-<td width="50%">
+### ⚡ Performance Optimizations
 
-### ⚡ Performance
-- **Flash Attention 2** — Auto-enabled
+- **Flash Attention 2** — Auto-enabled for speed
 - **torch.compile** — 2x faster training
 - **Dynamic Padding** — 50% less VRAM
 - **Triton Kernels** — Fused operations
 
-</td>
-</tr>
-<tr>
-<td>
-
 ### 🧠 45+ Model Architectures
-Llama 2/3, Mistral, Mixtral, Qwen 1/2, Phi 1/2/3, Gemma, Falcon, DeepSeek, Yi, StarCoder, ChatGLM, InternLM, Baichuan, StableLM, BLOOM, OPT, MPT, GPT-NeoX...
 
-</td>
-<td>
+Llama 2/3, Mistral, Mixtral, Qwen 1/2, Phi 1/2/3, Gemma, Falcon, DeepSeek, Yi, StarCoder, ChatGLM, InternLM, Baichuan, StableLM, BLOOM, OPT, MPT, GPT-NeoX...
 
 ### 📦 Multi-Format Export
-- **GGUF** — llama.cpp, Ollama, LM Studio
-- **ONNX** — ONNX Runtime, TensorRT
-- **MLX** — Apple Silicon (M1/M2/M3/M4)
-- **SafeTensors** — HuggingFace
 
-</td>
-</tr>
-<tr>
-<td>
+| Format | Use Case | Command |
+|--------|----------|---------|
+| **GGUF** | llama.cpp, Ollama, LM Studio | `model.export("gguf")` |
+| **ONNX** | ONNX Runtime, TensorRT | `model.export("onnx")` |
+| **MLX** | Apple Silicon (M1/M2/M3/M4) | `model.export("mlx")` |
+| **SafeTensors** | HuggingFace | `model.export("safetensors")` |
+
+### 🎨 Beautiful Console UI
 
-### 🎨 Beautiful UI
 ```
-╔════════════════════════════════════╗
-║  🚀 QuantLLM v2.0                  ║
-║  ✓ GGUF  ✓ ONNX  ✓ MLX             ║
-╚════════════════════════════════════╝
+╔════════════════════════════════════════════════════════════╗
+║   🚀 QuantLLM v2.0.0                                       ║
+║   Ultra-fast LLM Quantization & Export                     ║
+║   ✓ GGUF  ✓ ONNX  ✓ MLX  ✓ SafeTensors                     ║
+╚════════════════════════════════════════════════════════════╝
 
 📊 Model: meta-llama/Llama-3.2-3B
    Parameters: 3.21B
    Memory: 6.4 GB → 1.9 GB (70% saved)
 ```
 
-</td>
-<td>
-
 ### 🤗 One-Click Hub Publishing
-```python
-# Auto-generates model cards with:
-# - YAML frontmatter
-# - Usage examples  
-# - "Use this model" button
 
-model.push("user/my-model", format="gguf")
-```
+Auto-generates model cards with YAML frontmatter, usage examples, and "Use this model" button:
 
-</td>
-</tr>
-</table>
+```python
+model.push("user/my-model", format="gguf", quantization="Q4_K_M")
+```
 
 ---
 
@@ -225,12 +179,12 @@ model.export("safetensors", "./model-hf/")
 
 | Type | Bits | Quality | Use Case |
 |------|------|---------|----------|
-| `Q2_K` | 2-bit | Low | Minimum size |
-| `Q3_K_M` | 3-bit | Fair | Very constrained |
-| `Q4_K_M` | 4-bit | Good | **Recommended** ⭐ |
-| `Q5_K_M` | 5-bit | High | Quality-focused |
-| `Q6_K` | 6-bit | Very High | Near-original |
-| `Q8_0` | 8-bit | Excellent | Best quality |
+| `Q2_K` | 2-bit | 🔴 Low | Minimum size |
+| `Q3_K_M` | 3-bit | 🟠 Fair | Very constrained |
+| `Q4_K_M` | 4-bit | 🟢 Good | **Recommended** ⭐ |
+| `Q5_K_M` | 5-bit | 🟢 High | Quality-focused |
+| `Q6_K` | 6-bit | 🔵 Very High | Near-original |
+| `Q8_0` | 8-bit | 🔵 Excellent | Best quality |
 
 ---
 
@@ -260,17 +214,15 @@ response = model.chat(messages)
 print(response)
 ```
 
-### Load GGUF Models from HuggingFace
+### Load GGUF Models
 
 ```python
 from quantllm import TurboModel
 
-# Load any GGUF model directly
 model = TurboModel.from_gguf(
     "TheBloke/Llama-2-7B-Chat-GGUF", 
     filename="llama-2-7b-chat.Q4_K_M.gguf"
 )
-
 print(model.generate("Hello!"))
 ```
 
@@ -281,10 +233,10 @@ from quantllm import turbo
 
 model = turbo("mistralai/Mistral-7B")
 
-# Simple — everything auto-configured
+# Simple training
 model.finetune("training_data.json", epochs=3)
 
-# Advanced — full control
+# Advanced configuration
 model.finetune(
     "training_data.json",
     epochs=5,
@@ -296,6 +248,7 @@ model.finetune(
 ```
 
 **Supported data formats:**
+
 ```json
 [
   {"instruction": "What is Python?", "output": "Python is..."},
@@ -320,18 +273,12 @@ model.push(
 )
 ```
 
-The model card includes:
-- ✅ Proper YAML frontmatter (`library_name`, `tags`, `base_model`)
-- ✅ Format-specific usage examples
-- ✅ "Use this model" button compatibility
-- ✅ Quantization details
-
 ---
 
 ## 💻 Hardware Requirements
 
-| Configuration | GPU VRAM | Models |
-|---------------|----------|--------|
+| Configuration | GPU VRAM | Recommended Models |
+|---------------|----------|-------------------|
 | 🟢 **Entry** | 6-8 GB | 1-7B (4-bit) |
 | 🟡 **Mid-Range** | 12-24 GB | 7-30B (4-bit) |
 | 🔴 **High-End** | 24-80 GB | 70B+ |
@@ -343,7 +290,7 @@ The model card includes:
 ## 📦 Installation Options
 
 ```bash
-# Basic installation
+# Basic
 pip install git+https://github.com/codewithdark-git/QuantLLM.git
 
 # With specific features
@@ -356,17 +303,16 @@ pip install "quantllm[full]"     # Everything
 
 ---
 
-## 🏗️ Architecture
+## 🏗️ Project Structure
 
 ```
 quantllm/
-├── core/                    # Core functionality
+├── core/                    # Core API
 │   ├── turbo_model.py      # TurboModel unified API
-│   ├── smart_config.py     # Auto-configuration
-│   └── export.py           # Universal exporter
+│   └── smart_config.py     # Auto-configuration
 ├── quant/                   # Quantization
 │   └── llama_cpp.py        # GGUF conversion
-├── hub/                     # HuggingFace integration
+├── hub/                     # HuggingFace
 │   ├── hub_manager.py      # Push/pull models
 │   └── model_card.py       # Auto model cards
 ├── kernels/                 # Custom kernels
@@ -402,12 +348,12 @@ MIT License — see [LICENSE](LICENSE) for details.
 
 <div align="center">
 
-  ### Made with 🧡 by [Dark Coder](https://github.com/codewithdark-git)
+### Made with 🧡 by [Dark Coder](https://github.com/codewithdark-git)
 
-  <a href="https://github.com/codewithdark-git/QuantLLM">⭐ Star on GitHub</a> •
-  <a href="https://github.com/codewithdark-git/QuantLLM/issues">🐛 Report Bug</a> •
-  <a href="https://github.com/sponsors/codewithdark-git">💖 Sponsor</a>
+[⭐ Star on GitHub](https://github.com/codewithdark-git/QuantLLM) •
+[🐛 Report Bug](https://github.com/codewithdark-git/QuantLLM/issues) •
+[💖 Sponsor](https://github.com/sponsors/codewithdark-git)
 
-  **Happy Quantizing! 🚀**
+**Happy Quantizing! 🚀**
 
 </div>