| | <!DOCTYPE html> |
| | <html lang="en"> |
| | <head> |
| | <meta charset="UTF-8"> |
| | <meta name="viewport" content="width=device-width, initial-scale=1.0"> |
| | <title>GLM-4.7 Flash Deployment Hub</title> |
| | <meta name="description" content="Deployment interface for GLM-4.7-Flash-Uncensored model"> |
| | |
| | |
| | <script src="https://cdn.tailwindcss.com"></script> |
| | |
| | |
| | <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.0/css/all.min.css"> |
| | |
| | |
| | <link href="https://fonts.googleapis.com/css2?family=Inter:wght@300;400;600;700&family=Fira+Code:wght@400;600&display=swap" rel="stylesheet"> |
| |
|
| | <style> |
| | :root { |
| | --primary-color: #6366f1; |
| | --accent-color: #8b5cf6; |
| | --dark-bg: #0f172a; |
| | --panel-bg: #1e293b; |
| | --text-color: #e2e8f0; |
| | --success-color: #10b981; |
| | --terminal-font: 'Fira Code', monospace; |
| | } |
| | |
| | body { |
| | background-color: var(--dark-bg); |
| | color: var(--text-color); |
| | font-family: 'Inter', sans-serif; |
| | line-height: 1.6; |
| | } |
| | |
| | .glass-panel { |
| | background: rgba(30, 41, 59, 0.7); |
| | backdrop-filter: blur(10px); |
| | border: 1px solid rgba(255, 255, 255, 0.1); |
| | box-shadow: 0 4px 6px -1px rgba(0, 0, 0, 0.5); |
| | } |
| | |
| | .text-gradient { |
| | background: linear-gradient(to right, #818cf8, #c084fc); |
| | -webkit-background-clip: text; |
| | -webkit-text-fill-color: transparent; |
| | } |
| | |
| | .code-block { |
| | font-family: var(--terminal-font); |
| | background-color: #000; |
| | border-radius: 8px; |
| | padding: 1rem; |
| | position: relative; |
| | overflow-x: auto; |
| | } |
| | |
| | .code-header { |
| | background-color: #2d3748; |
| | padding: 0.5rem 1rem; |
| | border-radius: 8px 8px 0 0; |
| | font-size: 0.85rem; |
| | color: #a0aec0; |
| | display: flex; |
| | justify-content: space-between; |
| | align-items: center; |
| | } |
| | |
| | .terminal-window { |
| | background-color: #0d1117; |
| | border: 1px solid #30363d; |
| | border-radius: 12px; |
| | overflow: hidden; |
| | box-shadow: 0 10px 15px -3px rgba(0, 0, 0, 0.7); |
| | } |
| | |
| | .terminal-header { |
| | background-color: #161b22; |
| | padding: 0.75rem; |
| | border-bottom: 1px solid #30363d; |
| | display: flex; |
| | gap: 8px; |
| | } |
| | |
| | .dot { width: 12px; height: 12px; border-radius: 50%; } |
| | .red { background-color: #ff5f56; } |
| | .yellow { background-color: #ffbd2e; } |
| | .green { background-color: #27c93f; } |
| | |
| | .terminal-body { |
| | padding: 1rem; |
| | font-family: var(--terminal-font); |
| | font-size: 0.9rem; |
| | color: #c9d1d9; |
| | min-height: 200px; |
| | } |
| | |
| | .cmd-text { color: #7ee787; } |
| | .path-text { color: #58a6ff; } |
| | .output-text { color: #d2a8ff; } |
| | .error-text { color: #ff7b72; } |
| | .success-text { color: #3fb950; } |
| | |
| | .loader { |
| | border: 4px solid #f3f3f3; |
| | border-top: 4px solid var(--primary-color); |
| | border-radius: 50%; |
| | width: 24px; |
| | height: 24px; |
| | animation: spin 1s linear infinite; |
| | } |
| | |
| | @keyframes spin { |
| | 0% { transform: rotate(0deg); } |
| | 100% { transform: rotate(360deg); } |
| | } |
| | |
| | .feature-card:hover { |
| | transform: translateY(-5px); |
| | transition: all 0.3s ease; |
| | } |
| | </style> |
| | </head> |
| | <body class="min-h-screen flex flex-col"> |
| |
|
| | |
| | <nav class="glass-panel sticky top-0 z-50 border-b border-indigo-500/30"> |
| | <div class="max-w-7xl mx-auto px-4 sm:px-6 lg:px-8"> |
| | <div class="flex items-center justify-between h-16"> |
| | <div class="flex items-center gap-3"> |
| | <i class="fa-solid fa-robot text-2xl text-indigo-400"></i> |
| | <div> |
| | <h1 class="text-xl font-bold tracking-tight"> |
| | GLM-4.7 <span class="text-indigo-400">Flash</span> Deployer |
| | </h1> |
| | <p class="text-xs text-gray-400">Uncensored Code & Chat Interface</p> |
| | </div> |
| | </div> |
| | |
| | <div class="hidden md:flex items-center gap-6"> |
| | <a href="#dashboard" class="text-sm font-medium text-gray-300 hover:text-white transition">Dashboard</a> |
| | <a href="#install" class="text-sm font-medium text-gray-300 hover:text-white transition">Installation</a> |
| | <a href="#cloud" class="text-sm font-medium text-gray-300 hover:text-white transition">Cloud Options</a> |
| | <a href="https://huggingface.co/DavidAU/GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE-Imatrix-MAX-GGUF" target="_blank" class="text-sm font-medium text-indigo-400 hover:text-indigo-300 flex items-center gap-2"> |
| | <i class="fa-brands fa-huggingface"></i> Model Hub |
| | </a> |
| | </div> |
| |
|
| | |
| | <a href="https://huggingface.co/spaces/akhaliq/anycoder" target="_blank" class="text-xs font-semibold text-gray-500 hover:text-indigo-400 transition flex items-center gap-1"> |
| | <i class="fa-solid fa-link"></i> Built with anycoder |
| | </a> |
| | </div> |
| | </div> |
| | </nav> |
| |
|
| | |
| | <main class="flex-grow container mx-auto px-4 py-8 space-y-12"> |
| |
|
| | |
| | <section id="dashboard" class="text-center space-y-6 py-10"> |
| | <div class="inline-block px-4 py-1 rounded-full bg-indigo-900/30 border border-indigo-500/30 text-indigo-300 text-sm font-medium mb-4"> |
| | Model: GLM-4.7-Flash-Uncensored-Heretic-NEO-CODE |
| | </div> |
| | <h2 class="text-4xl md:text-6xl font-extrabold tracking-tight leading-tight"> |
| | Run the <span class="text-gradient">Ultimate</span> Local AI |
| | </h2> |
| | <p class="text-xl text-gray-400 max-w-3xl mx-auto"> |
| | A comprehensive dashboard to install, configure, and deploy the GLM-4.7 Flash model. |
| | Optimized for GGUF quantization to run efficiently on consumer hardware or cloud instances. |
| | </p> |
| |
|
| | <div class="flex flex-wrap justify-center gap-4 pt-4"> |
| | <a href="#install" class="px-8 py-3 rounded-lg bg-indigo-600 hover:bg-indigo-700 text-white font-bold shadow-lg shadow-indigo-500/30 transition transform hover:scale-105 flex items-center gap-2"> |
| | <i class="fa-solid fa-download"></i> Start Installation |
| | </a> |
| | <a href="#cloud" class="px-8 py-3 rounded-lg bg-gray-800 hover:bg-gray-700 text-white font-bold border border-gray-600 transition transform hover:scale-105 flex items-center gap-2"> |
| | <i class="fa-solid fa-cloud"></i> Cloud Deployment |
| | </a> |
| | </div> |
| | </section> |
| |
|
| | |
| | <section class="grid md:grid-cols-3 gap-6"> |
| | <div class="glass-panel p-6 rounded-xl feature-card"> |
| | <div class="w-12 h-12 rounded-lg bg-blue-900/50 flex items-center justify-center mb-4 text-blue-400"> |
| | <i class="fa-solid fa-memory text-2xl"></i> |
| | </div> |
| | <h3 class="text-lg font-bold mb-2">RAM Requirements</h3> |
| | <p class="text-sm text-gray-400"> |
| | The <strong>Imatrix-MAX</strong> version is large. We recommend at least <span class="text-green-400 font-bold">32GB RAM</span> for smooth inference without swapping. |
| | </p> |
| | </div> |
| | <div class="glass-panel p-6 rounded-xl feature-card"> |
| | <div class="w-12 h-12 rounded-lg bg-purple-900/50 flex items-center justify-center mb-4 text-purple-400"> |
| | <i class="fa-solid fa-microchip text-2xl"></i> |
| | </div> |
| | <h3 class="text-lg font-bold mb-2">GPU Recommendation</h3> |
| | <p class="text-sm text-gray-400"> |
| | An NVIDIA GPU with <span class="text-green-400 font-bold">8GB+ VRAM</span> (RTX 3060 or better) allows for acceleration via CUDA. |
| | </p> |
| | </div> |
| | <div class="glass-panel p-6 rounded-xl feature-card"> |
| | <div class="w-12 h-12 rounded-lg bg-emerald-900/50 flex items-center justify-center mb-4 text-emerald-400"> |
| | <i class="fa-solid fa-hard-drive text-2xl"></i> |
| | </div> |
| | <h3 class="text-lg font-bold mb-2">Storage</h3> |
| | <p class="text-sm text-gray-400"> |
| | The GGUF model typically ranges between <span class="text-green-400 font-bold">20GB - 40GB</span> depending on the specific quantization (Q4_K_M vs Q8_0). |
| | </p> |
| | </div> |
| | </section> |
| |
|
| | |
| | <section id="install" class="glass-panel rounded-2xl overflow-hidden"> |
| | <div class="p-6 border-b border-gray-700 bg-gray-900/50 flex justify-between items-center"> |
| | <div> |
| | <h3 class="text-xl font-bold flex items-center gap-2"> |
| | <i class="fa-solid fa-code"></i> Automated Setup Script |
| | </h3> |
| | <p class="text-sm text-gray-400">Generates the command line instructions for KoboldCpp & Ollama</p> |
| | </div> |
| | <button onclick="copyToClipboard()" id="copyBtn" class="px-4 py-2 rounded bg-gray-700 hover:bg-gray-600 text-white text-sm transition flex items-center gap-2"> |
| | <i class="fa-regular fa-copy"></i> Copy to Clipboard |
| | </button> |
| | </div> |
| | |
| | <div class="p-8 grid lg:grid-cols-2 gap-8"> |
| | |
| | <div class="space-y-6"> |
| | <h4 class="text-lg font-semibold text-indigo-400 border-b border-gray-700 pb-2">Configuration</h4> |
| | |
| | <div> |
| | <label class="block text-sm font-medium text-gray-300 mb-2">Execution Backend</label> |
| | <select id="backendSelect" class="w-full bg-gray-800 border border-gray-700 rounded-lg p-3 text-white focus:ring-2 focus:ring-indigo-500 focus:outline-none"> |
| | <option value="kobold">KoboldCpp (Local Web UI)</option> |
| | <option value="ollama">Ollama (System Service)</option> |
| | <option value="cmd">Direct CLI (llama.cpp)</option> |
| | </select> |
| | </div> |
| |
|
| | <div> |
| | <label class="block text-sm font-medium text-gray-300 mb-2">GPU Acceleration</label> |
| | <div class="flex gap-4"> |
| | <label class="flex items-center gap-2 text-gray-300 cursor-pointer"> |
| | <input type="radio" name="gpu" value="cpu" checked class="text-indigo-600 focus:ring-indigo-500"> |
| | <span>CPU Only (Slower)</span> |
| | </label> |
| | <label class="flex items-center gap-2 text-gray-300 cursor-pointer"> |
| | <input type="radio" name="gpu" value="cuda" class="text-indigo-600 focus:ring-indigo-500"> |
| | <span>NVIDIA (CUDA)</span> |
| | </label> |
| | <label class="flex items-center gap-2 text-gray-300 cursor-pointer"> |
| | <input type="radio" name="gpu" value="metal" class="text-indigo-600 focus:ring-indigo-500"> |
| | <span>Apple Silicon (Metal)</span> |
| | </label> |
| | </div> |
| | </div> |
| |
|
| | <div> |
| | <label class="block text-sm font-medium text-gray-300 mb-2">Threads (CPU Cores)</label> |
| | <input type="range" id="threadSlider" min="1" max="32" value="8" class="w-full h-2 bg-gray-700 rounded-lg appearance-none cursor-pointer accent-indigo-500"> |
| | <div class="flex justify-between text-xs text-gray-500 mt-1"> |
| | <span>1 Thread</span> |
| | <span id="threadValue" class="text-indigo-400 font-bold">8 Threads</span> |
| | <span>32 Threads</span> |
| | </div> |
| | </div> |
| |
|
| | <div class="bg-indigo-900/20 p-4 rounded-lg border border-indigo-500/20"> |
| | <h5 class="text-sm font-bold text-indigo-300 mb-2"><i class="fa-solid fa-circle-info"></i> Note on Imatrix-MAX</h5> |
| | <p class="text-xs text-gray-400"> |
| | The "Imatrix-MAX" version implies an optimized quantization matrix. Ensure you download the specific <code>.gguf</code> file from the HuggingFace link provided. The script below assumes standard GGUF loading. |
| | </p> |
| | </div> |
| | </div> |
| |
|
| | |
| | <div class="relative"> |
| | <div class="terminal-window"> |
| | <div class="terminal-header"> |
| | <div class="dot red"></div> |
| | <div class="dot yellow"></div> |
| | <div class="dot green"></div> |
| | <span class="text-xs text-gray-400 ml-2">root@server:~/glm-deploy</span> |
| | </div> |
| | <div class="terminal-body" id="codeOutput"> |
| | <div class="mb-2"> |
| | <span class="path-text">~</span> |
| | <span class="text-white"> $ </span> |
| | <span class="cmd-text"># Initializing GLM-4.7 Flash Setup...</span> |
| | </div> |
| | <div id="scriptContent"> |
| | <div class="mb-2"> |
| | <span class="path-text">~</span> |
| | <span class="text-white"> $ </span> |
| | <span class="cmd-text">git clone https://github.com/ggerganov/llama.cpp</span> |
| | </div> |
| | <div class="mb-2"> |
| | <span class="path-text">~</span> |
| | <span class="text-white"> $ </span> |
| | <span class="cmd-text">cd llama.cpp && make</span> |
| | </div> |
| | <div class="mb-2"> |
| | <span class="path-text">~/llama.cpp</span> |
| | <span class="text-white"> $ </span> |
| | <span class="cmd-text">./llama-server</span> |
| | <span class="text-gray-500"> \</span> |
| | </div> |
| | <div class="mb-2 ml-4"> |
| | <span class="text-gray-500"> --model ./models/glm-4.7-flash-uncensored.Q4_K_M.gguf</span> |
| | </div> |
| | <div class="mb-2 ml-4"> |
| | <span class="text-gray-500"> --n_ctx 4096 --n_gpu_layers 35</span> |
| | </div> |
| | <div class="mb-2 ml-4"> |
| | <span class="text-gray-500"> --port 8080 --host 0.0.0.0</span> |
| | </div> |
| | <div class="mt-4"> |
| | <span class="success-text">[INFO] Server starting on http://localhost:8080</span> |
| | </div> |
| | </div> |
| | </div> |
| | </div> |
| | <div class="mt-4 text-xs text-gray-500 text-center"> |
| | * Adjust <code>--n_gpu_layers</code> based on your VRAM capacity. |
| | </div> |
| | </div> |
| | </div> |
| | </section> |
| |
|
| | |
| | <section id="cloud" class="space-y-8"> |
| | <div class="text-center"> |
| | <h2 class="text-3xl font-bold">Cloud Deployment Options</h2> |
| | <p class="text-gray-400 mt-2">Running this model in the cloud requires instances with high RAM and VRAM.</p> |
| | </div> |
| |
|
| | <div class="grid md:grid-cols-3 gap-6"> |
| | |
| | <div class="glass-panel p-6 rounded-xl border-t-4 border-pink-500 flex flex-col justify-between hover:border-pink-400 transition"> |
| | <div> |
| | <div class="w-10 h-10 rounded bg-pink-900/30 flex items-center justify-center text-pink-400 mb-4"> |
| | <i class="fa-solid fa-cloud-bolt"></i> |
| | </div> |
| | <h3 class="text-xl font-bold mb-2">RunPod</h3> |
| | <p class="text-sm text-gray-400 mb-4"> |
| | Best for short bursts. Use an A100 or H100 pod. Upload your GGUF file to Pod Storage and run the server. |
| | </p> |
| | <div class="bg-gray-900/50 p-3 rounded text-xs font-mono text-pink-300 mb-4"> |
| | nvidia-a100-80gb |
| | </div> |
| | <div class="flex items-center gap-2 text-xs text-gray-500"> |
| | <i class="fa-solid fa-check text-green-500"></i> Pay per second |
| | </div> |
| | </div> |
| | <a href="https://runpod.io" target="_blank" class="mt-4 w-full text-center py-2 rounded bg-pink-600 hover:bg-pink-700 text-white text-sm font-medium transition"> |
| | Launch Pod |
| | </a> |
| | </div> |
| |
|
| | |
| | <div class="glass-panel p-6 rounded-xl border-t-4 border-orange-500 flex flex-col justify-between hover:border-orange-400 transition"> |
| | <div> |
| | <div class="w-10 h-10 rounded bg-orange-900/30 flex items-center justify-center text-orange-400 mb-4"> |
| | <i class="fa-solid fa-server"></i> |
| | </div> |
| | <h3 class="text-xl font-bold mb-2">Vast.ai</h3> |
| | <p class="text-sm text-gray-400 mb-4"> |
| | Marketplaces for GPU instances. Look for "RTX 4090" or "A100 80GB" instances. Very cost-effective. |
| | </p> |
| | <div class="bg-gray-900/50 p-3 rounded text-xs font-mono text-orange-300 mb-4"> |
| | 96GB RAM + 24GB VRAM |
| | </div> |
| | <div class="flex items-center gap-2 text-xs text-gray-500"> |
| | <i class="fa-solid fa-check text-green-500"></i> Bid/On-demand options |
| | </div> |
| | </div> |
| | <a href="https://vast.ai" target="_blank" class="mt-4 w-full text-center py-2 rounded bg-orange-600 hover:bg-orange-700 text-white text-sm font-medium transition"> |
| | Browse Instances |
| | </a> |
| | </div> |
| |
|
| | |
| | <div class="glass-panel p-6 rounded-xl border-t-4 border-blue-500 flex flex-col justify-between hover:border-blue-400 transition"> |
| | <div> |
| | <div class="w-10 h-10 rounded bg-blue-900/30 flex items-center justify-center text-blue-400 mb-4"> |
| | <i class="fa-solid faRocket"></i> |
| | </div> |
| | <h3 class="text-xl font-bold mb-2">Lambda Labs</h3> |
| | <p class="text-sm text-gray-400 mb-4"> |
| | User-friendly interface. Good for A10s and H100s. Easy to set up SSH and transfer models. |
| | </p> |
| | <div class="bg-gray-900/50 p-3 rounded text-xs font-mono text-blue-300 mb-4"> |
| | H100 80GB or A100 40/80GB |
| | </div> |
| | <div class="flex items-center gap-2 text-xs text-gray-500"> |
| | <i class="fa-solid fa-check text-green-500"></i> Flat rate pricing |
| | </div> |
| | </div> |
| | <a href="https://lambdalabs.com" target="_blank" class="mt-4 w-full text-center py-2 rounded bg-blue-600 hover:bg-blue-700 text-white text-sm font-medium transition"> |
| | Get API Key |
| | </a> |
| | </div> |
| | </div> |
| |
|
| | <div class="glass-panel p-6 rounded-xl bg-gradient-to-r from-gray-900 to-gray-800 border border-gray-700"> |
| | <h3 class="text-lg font-bold text-white mb-4"><i class="fa-solid fa-list-check"></i> Quick Cloud Setup Checklist</h3> |
| | <ul class="space-y-3"> |
| | <li class="flex items-start gap-3"> |
| | <div class="mt-1 w-5 h-5 rounded-full bg-green-900 flex items-center justify-center text-green-400 text-xs"><i class="fa-solid fa-check"></i></div> |
| | <div> |
| | <span class="font-bold text-gray-200">Select Instance:</span> |
| | <span class="text-gray-400 text-sm"> Ensure instance has > 32GB System RAM AND > 16GB VRAM for the Imatrix-MAX version.</span> |
| | </div> |
| | </li> |
| | <li class="flex items-start gap-3"> |
| | <div class="mt-1 w-5 h-5 rounded-full bg-green-900 flex items-center justify-center text-green-400 text-xs"><i class="fa-solid fa-check"></i></div> |
| | <div> |
| | <span class="font-bold text-gray-200">Download Model:</span> |
| | <span class="text-gray-400 text-sm"> Use `wget` or `huggingface-cli` on the cloud instance to download the GGUF file directly to storage.</span> |
| | </div> |
| | </li> |
| | <li class="flex items-start gap-3"> |
| | <div class="mt-1 w-5 h-5 rounded-full bg-green-900 flex items-center justify-center text-green-400 text-xs"><i class="fa-solid fa-check"></i></div> |
| | <div> |
| | <span class="font-bold text-gray-200">Start Server:</span> |
| | <span class="text-gray-400 text-sm"> Run the `llama-server` command with `--host 0.0.0.0` to allow external web access.</span> |
| | </div> |
| | </li> |
| | </ul> |
| | </div> |
| | </section> |
| |
|
| | |
| | <footer class="border-t border-gray-800 pt-8 pb-4 text-center"> |
| | <div class="flex items-center justify-center gap-2 mb-4 opacity-70"> |
| | <i class="fa-solid fa-robot text-indigo-500"></i> |
| | <span class="font-semibold text-gray-400">GLM-4.7 Flash Deployer</span> |
| | </div> |
| | <p class="text-sm text-gray-600"> |
| | This tool is a UI helper for deploying open-source AI models. |
| | Please ensure you comply with the model's license agreement. |
| | </p> |
| | <div class="mt-4 text-xs text-gray-700"> |
| | © <span id="year"></span> AI Deployment Tools. All rights reserved. |
| | </div> |
| | </footer> |
| |
|
| | </main> |
| |
|
| | |
| | <script> |
| | // Update Year |
| | document.getElementById('year').textContent = new Date().getFullYear(); |
| | |
| | // Thread Slider Logic |
| | const slider = document.getElementById('threadSlider'); |
| | const threadValue = document.getElementById('threadValue'); |
| | |
| | slider.addEventListener('input', function() { |
| | threadValue.textContent = this.value + " Threads"; |
| | updateScript(); |
| | }); |
| | |
| | // Backend Selection Logic |
| | const backendSelect = document.getElementById('backendSelect'); |
| | |
| | backendSelect.addEventListener('change', updateScript); |
| | |
| | // GPU Selection Logic |
| | const gpuRadios = document.querySelectorAll('input[name="gpu"]'); |
| | gpuRadios.forEach(radio => { |
| | radio.addEventListener('change', updateScript); |
| | }); |
| | |
| | function updateScript() { |
| | const backend = backendSelect.value; |
| | const threads = slider.value; |
| | const gpu = document.querySelector('input[name="gpu"]:checked').value; |
| | const outputDiv = document.getElementById('scriptContent'); |
| | |
| | let scriptHTML = ''; |
| | |
| | if (backend === 'kobold') { |
| | scriptHTML = ` |
| | <div class="mb-2"> |
| | <span class="path-text">~</span> |
| | <span class="text-white"> $ </span> |
| | <span class="cmd-text">git clone https://github.com/LostRuins/koboldcpp.git</span> |
| | </div> |
| | <div class="mb-2"> |
| | <span class="path-text">~</span> |
| | <span class="text-white"> $ </span> |
| | <span class="cmd-text">cd koboldcpp</span> |
| | </div> |
| | <div class="mb-2"> |
| | <span class="path-text">~/koboldcpp</span> |
| | <span class="text-white"> $ </span> |
| | <span class="cmd-text">mkdir -p models</span> |
| | </div> |
| | <div class="mb-2"> |
| | <span class="path-text">~/koboldcpp/models</span> |
| | <span class="text-white"> $ </span> |
| | <span class="cmd-text"># Place your GLM-4.7-Flash-Q4_K_M.gguf here</span> |
| | </div> |
| | <div class="mb-2"> |
| | <span class="path-text">~/koboldcpp</span> |
| | <span class="text-white"> $ </span> |
| | <span class="cmd-text">python3 koboldcpp.py</span> |
| | <span class="text-gray-500"> \</span> |
| | </div> |
| | <div class="mb-2 ml-4"> |
| | <span class="text-gray-500"> --model models/glm-4.7-flash-uncensored.Q4_K_M.gguf</span> |
| | </div> |
| | <div class="mb-2 ml-4"> |
| | <span class="text-gray-500"> --contextsize ${threads * 512} --threads ${threads}</span> |
| | </div> |
| | <div class="mb-2 ml-4"> |
| | <span class="text-gray-500"> --port 5001 --host 0.0.0.0</span> |
| | </div> |
| | <div class="mt-4"> |
| | <span class="success-text">[INFO] KoboldAI UI ready at http://localhost:5001</span> |
| | </div> |
| | `; |
| | } |
| | else if (backend === 'ollama') { |
| | scriptHTML = ` |
| | <div class="mb-2"> |
| | <span class="path-text">~</span> |
| | <span class="text-white"> $ </span> |
| | <span class="cmd-text">curl -fsSL https://ollama.com/install.sh | sh</span> |
| | </div> |
| | <div class="mb-2"> |
| | <span class="path-text">~</span> |
| | <span class="text-white"> $ </span> |
| | <span class="cmd-text">ollama pull nomic-ai/gpt4all-falcon</span> |
| | <span class="text-gray-500">/* Fallback if direct GLM pull fails */</span> |
| | </div> |
| | <div class="mb-2"> |
| | <span class="path-text">~</span> |
| | <span class="text-white"> $ </span> |
| | <span class="cmd-text">ollama create glm-4-flash -f Modelfile</span> |
| | </div> |
| | <div class="mb-2 bg-gray-800 p-3 rounded border-l-4 border-yellow-500"> |
| | <span class="text-xs text-gray-400 block mb-1"># Create a Modelfile:</span> |
| | <span class="text-xs text-white">FROM glm-4.7-flash-uncensored.Q4_K_M.gguf</span> |
| | <span class="text-xs text-white">PARAMETER num_thread ${threads}</span> |
| | </div> |
| | <div class="mt-4"> |
| | <span class="success-text">[SUCCESS] Model loaded into Ollama. Run 'ollama run glm-4-flash'</span> |
| | </div> |
| | `; |
| | } |
| | else { |
| | // Direct CLI |
| | scriptHTML = ` |
| | <div class="mb-2"> |
| | <span class="path-text">~</span> |
| | <span class="text-white"> $ </span> |
| | <span class="cmd-text">git clone https://github.com/ggerganov/llama.cpp</span> |
| | </div> |
| | <div class="mb-2"> |
| | <span class="path-text">~</span> |
| | <span class="text-white"> $ </span> |
| | <span class="cmd-text">cd llama.cpp && make</span> |
| | </div> |
| | <div class="mb-2"> |
| | <span class="path-text">~/llama.cpp</span> |
| | <span class="text-white"> $ </span> |
| | <span class="cmd-text">mkdir -p models</span> |
| | </div> |
| | <div class="mb-2"> |
| | <span class="path-text">~/llama.cpp/models</span> |
| | <span class="text-white"> $ </span> |
| | <span class="cmd-text"># Download GLM-4.7-Flash-Q4_K_M.gguf here</span> |
| | </div> |
| | <div class="mb-2"> |
| | <span class="path-text">~/llama.cpp</span> |
| | <span class="text-white"> $ </span> |
| | <span class="cmd-text">./main</span> |
| | <span class="text-gray-500"> \</span> |
| | </div> |
| | <div class="mb-2 ml-4"> |
| | <span class="text-gray-500"> -m models/glm-4.7-flash-uncensored.Q4_K_M.gguf</span> |
| | </div> |
| | <div class="mb-2 ml-4"> |
| | <span class="text-gray-500"> -n 512 --repeat_penalty 1.1</span> |
| | </div> |
| | <div class="mb-2 ml-4"> |
| | <span class="text-gray-500"> -c ${threads * 512} -t ${threads} -b 1</span> |
| | </div> |
| | <div class="mt-4"> |
| | <span class="success-text">[INFO] Entering interactive mode. Type 'exit' to quit.</span> |
| | </div> |
| | `; |
| | } |
| | |
| | // GPU Specific Flags |
| | if (gpu === 'cuda') { |
| | // Insert GPU flag logic roughly by appending to relevant lines or adding a new block |
| | // For simplicity in this template, we just update the text in the script |
| | const tempDiv = document.createElement('div'); |
| | tempDiv.innerHTML = scriptHTML; |
| | // Simple replacement logic based on backend would be complex, so we stick to generic output |
| | // In a real app, we'd inject --n_gpu_layers based on backend |
| | if (backend === 'kobold') { |
| | scriptHTML = scriptHTML.replace('--port 5001', '--port 5001 --gpu_layer_count 35'); |
| | } else if (backend === 'cmd') { |
| | scriptHTML = scriptHTML.replace('-t ' + threads, `-t ${threads} -ngl 35`); |
| | } |
| | } |
| | |
| | outputDiv.innerHTML = scriptHTML; |
| | } |
| | |
| | function copyToClipboard() { |
| | const codeElement = document.getElementById('scriptContent'); |
| | const textToCopy = codeElement.innerText; |
| | |
| | navigator.clipboard.writeText(textToCopy).then(() => { |
| | const btn = document.getElementById('copyBtn'); |
| | const originalText = btn.innerHTML; |
| | btn.innerHTML = '<i class="fa-solid fa-check"></i> Copied!'; |
| | btn.classList.remove('bg-gray-700'); |
| | btn.classList.add('bg-green-600'); |
| | |
| | setTimeout(() => { |
| | btn.innerHTML = originalText; |
| | btn.classList.add('bg-gray-700'); |
| | btn.classList.remove('bg-green-600'); |
| | }, 3000); |
| | }).catch(err => { |
| | console.error('Failed to copy: ', err); |
| | }); |
| | } |
| | </script> |
| | </body> |
| | </html> |
