Upload folder using huggingface_hub

README.md

@@ -1,10 +1,73 @@
 ---
-title: Qwen3.8 27B Uncensored MLX Demo
-emoji: 📊
-colorFrom: yellow
-colorTo: pink
+title: Qwen3.8-27B-Uncensored-MLX
+emoji: 🦙
+colorFrom: purple
+colorTo: indigo
 sdk: static
-pinned: false
+short_description: MLX demo for Qwen3.8-27B-Uncensored on Apple Silicon
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# Qwen3.8-27B-Uncensored-MLX Demo
+
+An interactive demo page for the [Qwen3.8-27B-Uncensored-MLX](https://huggingface.co/orcarouter/Qwen3.8-27B-Uncensored-MLX) model — an abliterated MLX build of Qwen's 27B vision-language model for Apple Silicon.
+
+## Live Chat Demo
+
+The embedded demo below uses the PyTorch equivalent on ZeroGPU. For native MLX inference on your Mac, see the [local usage instructions](#local-usage).
+
+## Model Overview
+
+**Qwen3.8-27B-Uncensored-MLX** is an abliterated (refusal-removed) MLX build of [`Qwen/Qwen3.8-27B`](https://huggingface.co/Qwen/Qwen3.8-27B) — a 27B-parameter dense, hybrid-attention native vision-language model with thinking control, tool-calling and an MTP head.
+
+### Key Features
+
+- **Abliterated**: Safety alignment removed for research purposes
+- **MLX Format**: Optimized for Apple Silicon (M1/M2/M3/M4)
+- **Vision-Language**: Native image understanding capabilities
+- **Multiple Quantizations**: 2 / 4 / 6 / 8-bit precision options
+- **Long Context**: 262K token context window
+
+### Available Quantizations
+
+| Precision | Size | Min Mac RAM | Quality |
+|-----------|------|-------------|---------|
+| 8-bit | ~27.5 GB | 32 GB | Near-lossless |
+| 6-bit | ~22 GB | 24-32 GB | Excellent |
+| 4-bit | ~15 GB | 24 GB | Very good (recommended) |
+| 2-bit | ~8.7 GB | 16 GB | Severely degraded |
+
+## Local Usage
+
+```bash
+# Install mlx-vlm
+pip install -U mlx-vlm
+
+# Download the 4-bit quantization
+hf download orcarouter/Qwen3.8-27B-Uncensored-MLX --include "4-bit/*" \
+    --local-dir ./Qwen3.8-27B-Uncensored-MLX
+
+# Text generation
+python -m mlx_vlm generate \
+    --model ./Qwen3.8-27B-Uncensored-MLX/4-bit \
+    --prompt "Explain quantum entanglement in one sentence." --max-tokens 256
+
+# Vision (image + text)
+python -m mlx_vlm generate \
+    --model ./Qwen3.8-27B-Uncensored-MLX/4-bit \
+    --image path/to/image.png \
+    --prompt "Describe this image." --max-tokens 256
+
+# OpenAI-compatible server
+python -m mlx_vlm server --model ./Qwen3.8-27B-Uncensored-MLX/4-bit --port 8080
+```
+
+## Links
+
+- **Model**: [orcarouter/Qwen3.8-27B-Uncensored-MLX](https://huggingface.co/orcarouter/Qwen3.8-27B-Uncensored-MLX)
+- **API**: [OrcaRouter Model API](https://www.orcarouter.ai/models/obsidian/qwen3.8-27b)
+- **Base Model**: [Qwen/Qwen3.8-27B](https://huggingface.co/Qwen/Qwen3.8-27B)
+- **Website**: [orcarouter.ai](https://www.orcarouter.ai)
+
+## Disclaimer
+
+This model has had its safety alignment substantially removed via abliteration. It is released strictly for legitimate research — interpretability, AI-safety and refusal-mechanism study, red-teaming, robustness evaluation, and controlled experiments. You assume full responsibility and liability for how you use it.

index.html

@@ -1,19 +1,396 @@
-<!doctype html>
-<html>
-	<head>
-		<meta charset="utf-8" />
-		<meta name="viewport" content="width=device-width" />
-		<title>My static Space</title>
-		<link rel="stylesheet" href="style.css" />
-	</head>
-	<body>
-		<div class="card">
-			<h1>Welcome to your static Space!</h1>
-			<p>You can modify this app directly by editing <i>index.html</i> in the Files and versions tab.</p>
-			<p>
-				Also don't forget to check the
-				<a href="https://huggingface.co/docs/hub/spaces" target="_blank">Spaces documentation</a>.
-			</p>
-		</div>
-	</body>
+<!DOCTYPE html>
+<html lang="en">
+<head>
+    <meta charset="UTF-8">
+    <meta name="viewport" content="width=device-width, initial-scale=1.0">
+    <title>Qwen3.8-27B-Uncensored-MLX Demo</title>
+    <style>
+        * {
+            margin: 0;
+            padding: 0;
+            box-sizing: border-box;
+        }
+        
+        body {
+            font-family: -apple-system, BlinkMacSystemFont, 'Segoe UI', Roboto, Oxygen, Ubuntu, sans-serif;
+            background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
+            min-height: 100vh;
+            color: #333;
+        }
+        
+        .container {
+            max-width: 1200px;
+            margin: 0 auto;
+            padding: 20px;
+        }
+        
+        header {
+            text-align: center;
+            padding: 40px 20px;
+            color: white;
+        }
+        
+        header h1 {
+            font-size: 2.5rem;
+            margin-bottom: 10px;
+            text-shadow: 0 2px 4px rgba(0,0,0,0.2);
+        }
+        
+        header p {
+            font-size: 1.1rem;
+            opacity: 0.9;
+            max-width: 700px;
+            margin: 0 auto;
+        }
+        
+        .badges {
+            display: flex;
+            justify-content: center;
+            gap: 10px;
+            margin-top: 20px;
+            flex-wrap: wrap;
+        }
+        
+        .badge {
+            background: rgba(255,255,255,0.2);
+            padding: 6px 14px;
+            border-radius: 20px;
+            font-size: 0.85rem;
+            backdrop-filter: blur(10px);
+        }
+        
+        .demo-section {
+            background: white;
+            border-radius: 16px;
+            padding: 24px;
+            margin: 20px 0;
+            box-shadow: 0 20px 60px rgba(0,0,0,0.3);
+        }
+        
+        .demo-section h2 {
+            color: #667eea;
+            margin-bottom: 16px;
+            font-size: 1.5rem;
+        }
+        
+        .demo-frame {
+            width: 100%;
+            height: 700px;
+            border: none;
+            border-radius: 12px;
+            background: #f5f5f5;
+        }
+        
+        .info-grid {
+            display: grid;
+            grid-template-columns: repeat(auto-fit, minmax(280px, 1fr));
+            gap: 20px;
+            margin: 20px 0;
+        }
+        
+        .info-card {
+            background: white;
+            border-radius: 12px;
+            padding: 24px;
+            box-shadow: 0 4px 20px rgba(0,0,0,0.1);
+        }
+        
+        .info-card h3 {
+            color: #764ba2;
+            margin-bottom: 12px;
+            font-size: 1.2rem;
+        }
+        
+        .info-card ul {
+            list-style: none;
+            padding: 0;
+        }
+        
+        .info-card li {
+            padding: 8px 0;
+            border-bottom: 1px solid #eee;
+            display: flex;
+            justify-content: space-between;
+        }
+        
+        .info-card li:last-child {
+            border-bottom: none;
+        }
+        
+        .info-card li span:first-child {
+            color: #666;
+        }
+        
+        .info-card li span:last-child {
+            font-weight: 600;
+            color: #333;
+        }
+        
+        .code-block {
+            background: #1e1e1e;
+            color: #d4d4d4;
+            padding: 20px;
+            border-radius: 12px;
+            overflow-x: auto;
+            font-family: 'SF Mono', Monaco, 'Cascadia Code', monospace;
+            font-size: 0.9rem;
+            line-height: 1.6;
+            margin: 16px 0;
+        }
+        
+        .code-block .comment {
+            color: #6a9955;
+        }
+        
+        .code-block .keyword {
+            color: #569cd6;
+        }
+        
+        .code-block .string {
+            color: #ce9178;
+        }
+        
+        .links {
+            display: flex;
+            justify-content: center;
+            gap: 16px;
+            margin: 30px 0;
+            flex-wrap: wrap;
+        }
+        
+        .link-btn {
+            background: white;
+            color: #667eea;
+            padding: 12px 24px;
+            border-radius: 8px;
+            text-decoration: none;
+            font-weight: 600;
+            transition: all 0.3s ease;
+            box-shadow: 0 4px 12px rgba(0,0,0,0.1);
+        }
+        
+        .link-btn:hover {
+            transform: translateY(-2px);
+            box-shadow: 0 8px 24px rgba(0,0,0,0.2);
+        }
+        
+        .disclaimer {
+            background: rgba(255, 193, 7, 0.1);
+            border: 1px solid rgba(255, 193, 7, 0.3);
+            border-radius: 12px;
+            padding: 20px;
+            margin: 20px 0;
+        }
+        
+        .disclaimer h4 {
+            color: #f57c00;
+            margin-bottom: 8px;
+        }
+        
+        .disclaimer p {
+            color: #666;
+            font-size: 0.9rem;
+            line-height: 1.6;
+        }
+        
+        footer {
+            text-align: center;
+            padding: 30px;
+            color: white;
+            opacity: 0.8;
+            font-size: 0.9rem;
+        }
+        
+        footer a {
+            color: white;
+            text-decoration: underline;
+        }
+        
+        .tabs {
+            display: flex;
+            gap: 8px;
+            margin-bottom: 16px;
+        }
+        
+        .tab {
+            padding: 8px 16px;
+            background: #f0f0f0;
+            border: none;
+            border-radius: 6px;
+            cursor: pointer;
+            font-size: 0.9rem;
+            transition: all 0.2s;
+        }
+        
+        .tab.active {
+            background: #667eea;
+            color: white;
+        }
+        
+        .tab-content {
+            display: none;
+        }
+        
+        .tab-content.active {
+            display: block;
+        }
+        
+        @media (max-width: 768px) {
+            header h1 {
+                font-size: 1.8rem;
+            }
+            
+            .demo-frame {
+                height: 500px;
+            }
+            
+            .info-grid {
+                grid-template-columns: 1fr;
+            }
+        }
+    </style>
+</head>
+<body>
+    <div class="container">
+        <header>
+            <h1>🦙 Qwen3.8-27B-Uncensored-MLX</h1>
+            <p>An abliterated MLX build of Qwen's 27B vision-language model for Apple Silicon. Text, images, and tool-calling with 262K context.</p>
+            <div class="badges">
+                <span class="badge">MLX Format</span>
+                <span class="badge">Apple Silicon</span>
+                <span class="badge">Vision-Language</span>
+                <span class="badge">262K Context</span>
+                <span class="badge">Abliterated</span>
+            </div>
+        </header>
+        
+        <div class="demo-section">
+            <h2>💬 Live Chat Demo</h2>
+            <p style="color: #666; margin-bottom: 16px;">
+                The demo below runs the PyTorch equivalent on ZeroGPU. For native MLX inference on your Mac, see the <a href="#local-usage" style="color: #667eea;">local usage section</a>.
+            </p>
+            <iframe 
+                class="demo-frame"
+                src="https://jonathancoletti-qwen3-8-27b-uncensored-demo.hf.space?__theme=light"
+                allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture"
+                allowfullscreen>
+            </iframe>
+        </div>
+        
+        <div class="info-grid">
+            <div class="info-card">
+                <h3>📊 Available Quantizations</h3>
+                <ul>
+                    <li><span>8-bit</span><span>~27.5 GB | 32 GB RAM</span></li>
+                    <li><span>6-bit</span><span>~22 GB | 24-32 GB RAM</span></li>
+                    <li><span>4-bit</span><span>~15 GB | 24 GB RAM</span></li>
+                    <li><span>2-bit</span><span>~8.7 GB | 16 GB RAM</span></li>
+                </ul>
+            </div>
+            
+            <div class="info-card">
+                <h3>✨ Key Features</h3>
+                <ul>
+                    <li><span>Architecture</span><span>Hybrid Attention</span></li>
+                    <li><span>Parameters</span><span>27 Billion</span></li>
+                    <li><span>Context</span><span>262K tokens</span></li>
+                    <li><span>Vision</span><span>Native VL Tower</span></li>
+                    <li><span>Tool Calling</span><span>Supported</span></li>
+                </ul>
+            </div>
+            
+            <div class="info-card">
+                <h3>🎯 Quality vs BF16</h3>
+                <ul>
+                    <li><span>8-bit</span><span>cos 0.9997</span></li>
+                    <li><span>6-bit</span><span>cos 0.9996</span></li>
+                    <li><span>4-bit</span><span>cos 0.996</span></li>
+                    <li><span>2-bit</span><span>cos 0.92 ⚠️</span></li>
+                </ul>
+            </div>
+        </div>
+        
+        <div class="demo-section" id="local-usage">
+            <h2>🚀 Local Usage (MLX on Apple Silicon)</h2>
+            <div class="tabs">
+                <button class="tab active" onclick="switchTab('install')">Install</button>
+                <button class="tab" onclick="switchTab('text')">Text</button>
+                <button class="tab" onclick="switchTab('vision')">Vision</button>
+                <button class="tab" onclick="switchTab('server')">Server</button>
+            </div>
+            
+            <div id="install" class="tab-content active">
+                <div class="code-block">
+<span class="comment"># Install mlx-vlm (requires mlx-vlm >= 0.6.13, mlx >= 0.32)</span>
+pip install -U mlx-vlm
+
+<span class="comment"># Download the 4-bit quantization (recommended)</span>
+hf download orcarouter/Qwen3.8-27B-Uncensored-MLX --include <span class="string">"4-bit/*"</span> \
+    --local-dir ./Qwen3.8-27B-Uncensored-MLX
+                </div>
+            </div>
+            
+            <div id="text" class="tab-content">
+                <div class="code-block">
+<span class="comment"># Text generation</span>
+python -m mlx_vlm generate \
+    --model ./Qwen3.8-27B-Uncensored-MLX/4-bit \
+    --prompt <span class="string">"Explain quantum entanglement in one sentence."</span> \
+    --max-tokens 256
+                </div>
+            </div>
+            
+            <div id="vision" class="tab-content">
+                <div class="code-block">
+<span class="comment"># Vision (image + text)</span>
+python -m mlx_vlm generate \
+    --model ./Qwen3.8-27B-Uncensored-MLX/4-bit \
+    --image path/to/image.png \
+    --prompt <span class="string">"Describe this image."</span> \
+    --max-tokens 256
+                </div>
+            </div>
+            
+            <div id="server" class="tab-content">
+                <div class="code-block">
+<span class="comment"># OpenAI-compatible server</span>
+python -m mlx_vlm server \
+    --model ./Qwen3.8-27B-Uncensored-MLX/4-bit \
+    --port 8080
+
+<span class="comment"># Then use with any OpenAI-compatible client</span>
+curl http://localhost:8080/v1/chat/completions \
+  -H <span class="string">"Content-Type: application/json"</span> \
+  -d <span class="string">'{"messages": [{"role": "user", "content": "Hello!"}]}'</span>
+                </div>
+            </div>
+        </div>
+        
+        <div class="disclaimer">
+            <h4>⚠️ Disclaimer</h4>
+            <p>This model has had its safety alignment substantially removed via abliteration. It is released strictly for legitimate research — interpretability, AI-safety and refusal-mechanism study, red-teaming, robustness evaluation, and controlled experiments. You assume full responsibility and liability for how you use it and for everything it generates.</p>
+        </div>
+        
+        <div class="links">
+            <a href="https://huggingface.co/orcarouter/Qwen3.8-27B-Uncensored-MLX" class="link-btn" target="_blank">📦 Model on HF</a>
+            <a href="https://www.orcarouter.ai/models/obsidian/qwen3.8-27b" class="link-btn" target="_blank">🔌 API Endpoint</a>
+            <a href="https://huggingface.co/Qwen/Qwen3.8-27B" class="link-btn" target="_blank">🏗️ Base Model</a>
+            <a href="https://www.orcarouter.ai" class="link-btn" target="_blank">🌐 OrcaRouter</a>
+        </div>
+        
+        <footer>
+            <p>Built with ❤️ for the open-source ML community</p>
+            <p>Model by <a href="https://huggingface.co/orcarouter">OrcaRouter</a> | Base by <a href="https://huggingface.co/Qwen">Qwen</a></p>
+        </footer>
+    </div>
+    
+    <script>
+        function switchTab(tabId) {
+            document.querySelectorAll('.tab-content').forEach(el => el.classList.remove('active'));
+            document.querySelectorAll('.tab').forEach(el => el.classList.remove('active'));
+            document.getElementById(tabId).classList.add('active');
+            event.target.classList.add('active');
+        }
+    </script>
+</body>
 </html>