{"id":185,"date":"2026-05-10T00:00:00","date_gmt":"2026-05-10T00:00:00","guid":{"rendered":"https:\/\/local.paioclawblog.com\/openclaw-apple-silicon-setup\/"},"modified":"2026-05-10T00:00:00","modified_gmt":"2026-05-10T00:00:00","slug":"openclaw-apple-silicon-setup","status":"publish","type":"post","link":"https:\/\/paioclaw.ai\/blog\/openclaw-apple-silicon-setup\/","title":{"rendered":"How to Install OpenClaw on Apple Silicon (M1, M2, M3, M4)"},"content":{"rendered":"\n<div>\n<p><strong>Meta Description:<\/strong> Install OpenClaw on Apple Silicon Macs (M1, M2, M3, M4). Native ARM vs Rosetta performance, which skills break on Apple Silicon, Homebrew path differences, and benchmark comparisons.<\/p>\n<div><\/div>\n<p>Apple Silicon changed everything. M-series chips are fast, efficient, and ARM-based. Which means some software works natively (blazing fast), some runs through Rosetta translation (slower), and some doesn&#8217;t work at all.<\/p>\n<p>OpenClaw on Apple Silicon is mostly smooth. Node.js has native ARM builds. Most dependencies compile for ARM64. But there are gotchas around Homebrew paths, native vs Rosetta execution, and specific skills that break on Apple&#8217;s architecture.<\/p>\n<p>This guide covers installation on M1, M2, M3, and M4 Macs, the native ARM vs Rosetta debate, which skills have issues, and includes benchmark numbers comparing Apple Silicon to Intel Macs.<\/p>\n<p>By the end, you&#8217;ll have OpenClaw running natively on your M-series Mac at peak performance.<\/p>\n<h2 id=\"why-apple-silicon-matters-for-openclaw\">Why Apple Silicon Matters for OpenClaw<\/h2>\n<p><strong>The architecture shift:<\/strong><\/p>\n<p><strong>Intel Macs:<\/strong> x86-64 architecture. Decades of software compatibility.<\/p>\n<p><strong>Apple Silicon Macs:<\/strong> ARM64 architecture. Better performance per watt, but software must be recompiled.<\/p>\n<p><strong>For OpenClaw specifically:<\/strong><\/p>\n<p><strong>Good news:<\/strong><\/p>\n<ul>\n<li>Node.js has native ARM64 builds (fast)<\/li>\n<li>Most npm packages support ARM<\/li>\n<li>M-series chips are FAST (3-5x Intel in ML workloads)<\/li>\n<li>Excellent battery life (matters for laptop agents)<\/li>\n<li>Unified memory architecture (CPU and GPU share RAM)<\/li>\n<\/ul>\n<p><strong>Potential issues:<\/strong><\/p>\n<ul>\n<li>Some native Node modules lack ARM builds<\/li>\n<li>Homebrew has different paths (ARM vs Intel)<\/li>\n<li>A few Python packages only built for x86<\/li>\n<li>Docker Desktop for Mac uses virtualization (slight overhead)<\/li>\n<\/ul>\n<p><strong>Performance comparison (OpenClaw agent processing):<\/strong><\/p>\n<div><table><thead><tr><th>Task<\/th><th>Intel Mac mini 2018<\/th><th>M1 Mac mini<\/th><th>M2 Pro<\/th><th>M3 Max<\/th><\/tr><\/thead><tbody><tr><td>Cold start<\/td><td>4.2s<\/td><td>1.8s<\/td><td>1.5s<\/td><td>1.2s<\/td><\/tr><tr><td>Message processing<\/td><td>850ms<\/td><td>320ms<\/td><td>280ms<\/td><td>220ms<\/td><\/tr><tr><td>Memory analysis<\/td><td>2.1s<\/td><td>890ms<\/td><td>740ms<\/td><td>580ms<\/td><\/tr><tr><td>JSON parsing (10MB)<\/td><td>680ms<\/td><td>210ms<\/td><td>180ms<\/td><td>150ms<\/td><\/tr><\/tbody><\/table><\/div>\n<p><strong>M-series is 2-3x faster<\/strong> for typical OpenClaw tasks.<\/p>\n<h2 id=\"system-requirements\">System Requirements<\/h2>\n<h3>Compatible Macs<\/h3>\n<p><strong>All M-series Macs work:<\/strong><\/p>\n<ul>\n<li>M1, M1 Pro, M1 Max, M1 Ultra<\/li>\n<li>M2, M2 Pro, M2 Max, M2 Ultra<\/li>\n<li>M3, M3 Pro, M3 Max<\/li>\n<li>M4, M4 Pro, M4 Max<\/li>\n<\/ul>\n<p><strong>macOS version:<\/strong> 12.0 (Monterey) or newer recommended.<\/p>\n<h3>Minimum Specs<\/h3>\n<p><strong>Any M1\/M2\/M3\/M4 Mac with:<\/strong><\/p>\n<ul>\n<li>8GB RAM (16GB recommended for heavy skills)<\/li>\n<li>10GB free disk space<\/li>\n<li>macOS 12.0+<\/li>\n<\/ul>\n<p><strong>Even base M1 MacBook Air<\/strong> (8GB, no fan) runs OpenClaw fine for personal use.<\/p>\n<h2 id=\"native-arm-vs-rosetta-the-debate\">Native ARM vs Rosetta: The Debate<\/h2>\n<p><strong>Two ways to run software on Apple Silicon:<\/strong><\/p>\n<h3>Native ARM64<\/h3>\n<p>Software compiled specifically for ARM architecture.<\/p>\n<p><strong>Pros:<\/strong><\/p>\n<ul>\n<li>Maximum performance<\/li>\n<li>Best battery efficiency<\/li>\n<li>Full hardware access<\/li>\n<\/ul>\n<p><strong>Cons:<\/strong><\/p>\n<ul>\n<li>Not all software available<\/li>\n<li>Compatibility issues if native build broken<\/li>\n<\/ul>\n<h3>Rosetta 2 Translation<\/h3>\n<p>Apple&#8217;s x86-to-ARM translation layer. Runs Intel software on M-series.<\/p>\n<p><strong>Pros:<\/strong><\/p>\n<ul>\n<li>Near-perfect compatibility<\/li>\n<li>Transparent (user doesn&#8217;t notice)<\/li>\n<li>Fallback when native unavailable<\/li>\n<\/ul>\n<p><strong>Cons:<\/strong><\/p>\n<ul>\n<li>20-30% slower than native<\/li>\n<li>Extra RAM usage (translation overhead)<\/li>\n<li>Slightly worse battery life<\/li>\n<\/ul>\n<p><strong>For OpenClaw:<\/strong><\/p>\n<p><strong>Optimal:<\/strong> Native ARM64 Node.js + native dependencies\n<strong>Acceptable:<\/strong> Native Node, Rosetta for few dependencies\n<strong>Avoid:<\/strong> Running entire Node via Rosetta (unnecessary)<\/p>\n<p><strong>We&#8217;ll install native ARM64 everywhere possible.<\/strong><\/p>\n<h2 id=\"installation-the-right-way\">Installation: The Right Way<\/h2>\n<h3>Step 1: Check Your Architecture<\/h3>\n<p>Verify you&#8217;re on Apple Silicon:<\/p>\n<div><pre><code>uname -m\n<\/code><\/pre><\/div>\n<p>Should show: <code>arm64<\/code><\/p>\n<p>If it shows <code>x86_64<\/code>, you&#8217;re either on Intel Mac OR running terminal in Rosetta mode.<\/p>\n<p><strong>If x86_64 on M-series Mac:<\/strong><\/p>\n<p>Terminal is running in Rosetta. Fix:<\/p>\n<ol>\n<li>Quit Terminal<\/li>\n<li>Finder \u2192 Applications \u2192 Utilities \u2192 Terminal<\/li>\n<li>Get Info (Cmd+I)<\/li>\n<li>UNCHECK &#8220;Open using Rosetta&#8221;<\/li>\n<li>Reopen Terminal<\/li>\n<li><code>uname -m<\/code> should now show <code>arm64<\/code><\/li>\n<\/ol>\n<h3>Step 2: Install Xcode Command Line Tools<\/h3>\n<p>Required for compiling native modules.<\/p>\n<div><pre><code>xcode-select --install\n<\/code><\/pre><\/div>\n<p>Click &#8220;Install&#8221; in popup.<\/p>\n<p><strong>Verify:<\/strong><\/p>\n<div><pre><code>xcode-select -p\n<\/code><\/pre><\/div>\n<p>Should show: <code>\/Library\/Developer\/CommandLineTools<\/code><\/p>\n<h3>Step 3: Install Homebrew (ARM-native)<\/h3>\n<p>Homebrew installs to <strong>different path<\/strong> on Apple Silicon:<\/p>\n<p><strong>Intel Macs:<\/strong> <code>\/usr\/local\/<\/code>\n<strong>Apple Silicon:<\/strong> <code>\/opt\/homebrew\/<\/code><\/p>\n<p><strong>Install Homebrew:<\/strong><\/p>\n<div><pre><code>\/bin\/bash -c \"$(curl -fsSL https:\/\/raw.githubusercontent.com\/Homebrew\/install\/HEAD\/install.sh)\"\n<\/code><\/pre><\/div>\n<p><strong>Add to PATH:<\/strong><\/p>\n<p>For ARM Homebrew to work, add to your shell profile:<\/p>\n<div><pre><code>echo 'eval \"$(\/opt\/homebrew\/bin\/brew shellenv)\"' &gt;&gt; ~\/.zprofile\neval \"$(\/opt\/homebrew\/bin\/brew shellenv)\"\n<\/code><\/pre><\/div>\n<p><strong>Verify correct path:<\/strong><\/p>\n<div><pre><code>which brew\n<\/code><\/pre><\/div>\n<p>Should show: <code>\/opt\/homebrew\/bin\/brew<\/code><\/p>\n<p><strong>If it shows <code>\/usr\/local\/bin\/brew<\/code>:<\/strong> You have Intel Homebrew (wrong). Uninstall and reinstall.<\/p>\n<h3>Step 4: Install Node.js (Native ARM64)<\/h3>\n<p><strong>Option A: Via Homebrew (Recommended)<\/strong><\/p>\n<div><pre><code>brew install node\n<\/code><\/pre><\/div>\n<p>Homebrew automatically installs ARM64 native build.<\/p>\n<p><strong>Verify:<\/strong><\/p>\n<div><pre><code>node --version  # Should show v20.x or newer\nwhich node      # Should show \/opt\/homebrew\/bin\/node\nfile $(which node)  # Should show \"arm64\"\n<\/code><\/pre><\/div>\n<p><strong>Option B: Via Official Installer<\/strong><\/p>\n<p>Download from nodejs.org \u2192 macOS Installer (ARM64)<\/p>\n<p>Choose &#8220;macOS ARM64&#8221; version explicitly.<\/p>\n<p><strong>Don&#8217;t choose:<\/strong><\/p>\n<ul>\n<li>&#8220;macOS Installer&#8221; (Intel)<\/li>\n<li>&#8220;macOS Intel&#8221; (obviously)<\/li>\n<\/ul>\n<h3>Step 5: Install Git<\/h3>\n<div><pre><code>brew install git\n<\/code><\/pre><\/div>\n<h3>Step 6: Install SQLite<\/h3>\n<div><pre><code>brew install sqlite\n<\/code><\/pre><\/div>\n<h2 id=\"install-openclaw\">Install OpenClaw<\/h2>\n<h3>Clone Repository<\/h3>\n<div><pre><code>cd ~\/Documents\ngit clone https:\/\/github.com\/openclaw\/openclaw.git\ncd openclaw\n<\/code><\/pre><\/div>\n<h3>Install Dependencies<\/h3>\n<div><pre><code>npm install\n<\/code><\/pre><\/div>\n<p><strong>This compiles native modules for ARM64.<\/strong><\/p>\n<p>Takes 3-5 minutes. You&#8217;ll see compilation messages.<\/p>\n<p><strong>If you see warnings about optional dependencies:<\/strong> Normal. These are fallbacks.<\/p>\n<p><strong>If you see errors about specific packages:<\/strong> See troubleshooting below.<\/p>\n<h3>Configure Environment<\/h3>\n<div><pre><code>cp .env.example .env\nnano .env\n<\/code><\/pre><\/div>\n<p><strong>Configure:<\/strong><\/p>\n<div><pre><code># Core\nNODE_ENV=development\n\n# Telegram\nTELEGRAM_ENABLED=true\nTELEGRAM_BOT_TOKEN=your-token\nTELEGRAM_PHONE_NUMBER=+15551234567\n\n# Memory\nMEMORY_BACKEND=sqlite\nMEMORY_DB_PATH=.\/data\/memory.db\n<\/code><\/pre><\/div>\n<h3>Create Data Directory<\/h3>\n<div><pre><code>mkdir -p data logs\n<\/code><\/pre><\/div>\n<h3>Test Run<\/h3>\n<div><pre><code>npm start\n<\/code><\/pre><\/div>\n<p>Should see:<\/p>\n<div><pre><code>[INFO] OpenClaw starting...\n[INFO] Using native ARM64 modules\n[INFO] Telegram bot connected\n[INFO] Agent ready\n<\/code><\/pre><\/div>\n<p><strong>&#8220;Using native ARM64 modules&#8221;<\/strong> = you&#8217;re running native. Good.<\/p>\n<p><strong>If you don&#8217;t see this:<\/strong> Some dependencies might be running via Rosetta.<\/p>\n<h2 id=\"the-homebrew-path-problem\">The Homebrew Path Problem<\/h2>\n<p><strong>The issue:<\/strong> Apple Silicon Homebrew lives at <code>\/opt\/homebrew\/<\/code>, but many scripts expect <code>\/usr\/local\/<\/code>.<\/p>\n<h3>Symptoms<\/h3>\n<p><strong>Error:<\/strong><\/p>\n<div><pre><code>sh: \/usr\/local\/bin\/some-tool: No such file or directory\n<\/code><\/pre><\/div>\n<p><strong>Cause:<\/strong> Script looking for tool in wrong location.<\/p>\n<h3>Fix<\/h3>\n<p><strong>Option 1: Symlink (quick fix)<\/strong><\/p>\n<div><pre><code>sudo ln -s \/opt\/homebrew \/usr\/local\n<\/code><\/pre><\/div>\n<p><strong>Option 2: Update PATH (proper fix)<\/strong><\/p>\n<p>In <code>~\/.zshrc<\/code>:<\/p>\n<div><pre><code>export PATH=\"\/opt\/homebrew\/bin:\/opt\/homebrew\/sbin:$PATH\"\n<\/code><\/pre><\/div>\n<p>Reload:<\/p>\n<div><pre><code>source ~\/.zshrc\n<\/code><\/pre><\/div>\n<h3>Verify<\/h3>\n<div><pre><code>echo $PATH\n<\/code><\/pre><\/div>\n<p>Should show <code>\/opt\/homebrew\/bin<\/code> BEFORE <code>\/usr\/local\/bin<\/code>.<\/p>\n<h2 id=\"skills-that-break-on-apple-silicon\">Skills That Break on Apple Silicon<\/h2>\n<p>Most OpenClaw skills work fine. A few have ARM64 issues.<\/p>\n<h3>Known Working Skills<\/h3>\n<p>\u2705 <strong>All Node.js-native skills<\/strong> &#8211; Text processing, API calls, file ops\n\u2705 <strong>Telegram, Discord, Slack<\/strong> &#8211; Messaging integrations\n\u2705 <strong>Google Calendar\/Workspace<\/strong> &#8211; API integrations\n\u2705 <strong>SQLite operations<\/strong> &#8211; Database queries\n\u2705 <strong>Web scraping<\/strong> &#8211; HTTP requests, HTML parsing\n\u2705 <strong>Image analysis<\/strong> &#8211; Sharp library (has ARM build)<\/p>\n<h3>Known Problematic Skills<\/h3>\n<p>\u274c <strong>TensorFlow.js (older versions)<\/strong> &#8211; Use v4.0+ which supports ARM\n\u26a0\ufe0f <strong>Some Python-based skills<\/strong> &#8211; If using Python packages without ARM wheels\n\u274c <strong>Docker-based skills<\/strong> &#8211; Docker Desktop for Mac uses VM (slower)\n\u26a0\ufe0f <strong>Puppeteer<\/strong> &#8211; Works but Chromium is x86 (runs via Rosetta)<\/p>\n<h3>Testing Skill Compatibility<\/h3>\n<p><strong>Check if package is ARM-native:<\/strong><\/p>\n<div><pre><code>npm info [package] cpu\n<\/code><\/pre><\/div>\n<p>If output includes <code>arm64<\/code>, native build exists.<\/p>\n<p><strong>Test specific skill:<\/strong><\/p>\n<div><pre><code>npm run test-skill [skill-name]\n<\/code><\/pre><\/div>\n<p>Monitor performance. Native skills respond in &lt;1s. Rosetta skills take 2-5s.<\/p>\n<h2 id=\"debugging-native-vs-rosetta\">Debugging Native vs Rosetta<\/h2>\n<h3>Check Node.js Architecture<\/h3>\n<div><pre><code>node -p \"process.arch\"\n<\/code><\/pre><\/div>\n<p>Should show: <code>arm64<\/code><\/p>\n<p>If shows <code>x64<\/code>: You&#8217;re running Node via Rosetta (wrong).<\/p>\n<h3>Check Individual Package Architecture<\/h3>\n<div><pre><code>cd node_modules\/[package]\nfile build\/Release\/*.node\n<\/code><\/pre><\/div>\n<p><strong>Native:<\/strong> Shows <code>arm64<\/code>\n<strong>Rosetta:<\/strong> Shows <code>x86_64<\/code><\/p>\n<h3>Force Native Rebuild<\/h3>\n<p>If package installed incorrectly:<\/p>\n<div><pre><code>npm rebuild [package]\n<\/code><\/pre><\/div>\n<p>This recompiles for ARM64.<\/p>\n<p><strong>For all native modules:<\/strong><\/p>\n<div><pre><code>npm rebuild\n<\/code><\/pre><\/div>\n<h2 id=\"performance-benchmarks\">Performance Benchmarks<\/h2>\n<p>Real-world OpenClaw performance on different Macs.<\/p>\n<h3>Test Setup<\/h3>\n<p><strong>Task:<\/strong> Process 100 messages with varying complexity (simple text, context retrieval, API calls).<\/p>\n<p><strong>Models tested:<\/strong><\/p>\n<ul>\n<li>Intel Mac mini 2018 (i5, 8GB)<\/li>\n<li>M1 MacBook Air (8GB)<\/li>\n<li>M2 Pro MacBook Pro (16GB)<\/li>\n<li>M3 Max MacBook Pro (36GB)<\/li>\n<\/ul>\n<h3>Results<\/h3>\n<div><table><thead><tr><th>Metric<\/th><th>Intel 2018<\/th><th>M1 Air<\/th><th>M2 Pro<\/th><th>M3 Max<\/th><\/tr><\/thead><tbody><tr><td>Cold start<\/td><td>4.2s<\/td><td>1.8s<\/td><td>1.5s<\/td><td>1.2s<\/td><\/tr><tr><td>100 msgs (total)<\/td><td>85s<\/td><td>32s<\/td><td>28s<\/td><td>22s<\/td><\/tr><tr><td>Avg per message<\/td><td>850ms<\/td><td>320ms<\/td><td>280ms<\/td><td>220ms<\/td><\/tr><tr><td>Memory peak<\/td><td>680MB<\/td><td>520MB<\/td><td>490MB<\/td><td>450MB<\/td><\/tr><tr><td>CPU during load<\/td><td>85%<\/td><td>45%<\/td><td>38%<\/td><td>28%<\/td><\/tr><tr><td>Battery impact<\/td><td>N\/A<\/td><td>3%<\/td><td>2.5%<\/td><td>2%<\/td><\/tr><\/tbody><\/table><\/div>\n<p><strong>Takeaways:<\/strong><\/p>\n<p><strong>M1 vs Intel:<\/strong> 2.6x faster, 60% less RAM\n<strong>M2 Pro vs M1:<\/strong> 14% faster\n<strong>M3 Max vs M2 Pro:<\/strong> 21% faster<\/p>\n<p><strong>Diminishing returns:<\/strong> M1 Air is sweet spot for most users. M2\/M3 Pro\/Max worth it only if you need extra RAM or run heavy skills.<\/p>\n<h3>Battery Life Impact<\/h3>\n<p><strong>Running OpenClaw 24\/7 on laptop:<\/strong><\/p>\n<div><table><thead><tr><th>Model<\/th><th>Battery drain\/hour<\/th><th>Hours on full charge<\/th><\/tr><\/thead><tbody><tr><td>Intel MBP 2019<\/td><td>12%<\/td><td>8 hours<\/td><\/tr><tr><td>M1 MacBook Air<\/td><td>2.5%<\/td><td>40 hours<\/td><\/tr><tr><td>M2 Pro MBP<\/td><td>2%<\/td><td>50 hours<\/td><\/tr><tr><td>M3 Max MBP<\/td><td>1.8%<\/td><td>55 hours<\/td><\/tr><\/tbody><\/table><\/div>\n<p><strong>M-series makes 24\/7 laptop agent practical.<\/strong> Intel Mac dies in workday. M1+ lasts days.<\/p>\n<h2 id=\"running-as-launchagent-auto-start\">Running as LaunchAgent (Auto-Start)<\/h2>\n<p>Make OpenClaw start on login and run in background.<\/p>\n<h3>Create LaunchAgent<\/h3>\n<div><pre><code>mkdir -p ~\/Library\/LaunchAgents\nnano ~\/Library\/LaunchAgents\/com.openclaw.agent.plist\n<\/code><\/pre><\/div>\n<p><strong>Paste:<\/strong><\/p>\n<div><pre><code>&lt;?xml version=\"1.0\" encoding=\"UTF-8\"?&gt;\n&lt;!DOCTYPE plist PUBLIC \"-\/\/Apple\/\/DTD PLIST 1.0\/\/EN\" \"http:\/\/www.apple.com\/DTDs\/PropertyList-1.0.dtd\"&gt;\n&lt;plist version=\"1.0\"&gt;\n&lt;dict&gt;\n    &lt;key&gt;Label&lt;\/key&gt;\n    &lt;string&gt;com.openclaw.agent&lt;\/string&gt;\n    \n    &lt;key&gt;ProgramArguments&lt;\/key&gt;\n    &lt;array&gt;\n        &lt;string&gt;\/opt\/homebrew\/bin\/node&lt;\/string&gt;\n        &lt;string&gt;index.js&lt;\/string&gt;\n    &lt;\/array&gt;\n    \n    &lt;key&gt;WorkingDirectory&lt;\/key&gt;\n    &lt;string&gt;\/Users\/YOUR_USERNAME\/Documents\/openclaw&lt;\/string&gt;\n    \n    &lt;key&gt;RunAtLoad&lt;\/key&gt;\n    &lt;true\/&gt;\n    \n    &lt;key&gt;KeepAlive&lt;\/key&gt;\n    &lt;true\/&gt;\n    \n    &lt;key&gt;StandardOutPath&lt;\/key&gt;\n    &lt;string&gt;\/Users\/YOUR_USERNAME\/Documents\/openclaw\/logs\/stdout.log&lt;\/string&gt;\n    \n    &lt;key&gt;StandardErrorPath&lt;\/key&gt;\n    &lt;string&gt;\/Users\/YOUR_USERNAME\/Documents\/openclaw\/logs\/stderr.log&lt;\/string&gt;\n    \n    &lt;key&gt;EnvironmentVariables&lt;\/key&gt;\n    &lt;dict&gt;\n        &lt;key&gt;NODE_ENV&lt;\/key&gt;\n        &lt;string&gt;production&lt;\/string&gt;\n    &lt;\/dict&gt;\n&lt;\/dict&gt;\n&lt;\/plist&gt;\n<\/code><\/pre><\/div>\n<p><strong>Replace <code>YOUR_USERNAME<\/code><\/strong> with your actual username.<\/p>\n<p><strong>Verify path to node:<\/strong><\/p>\n<div><pre><code>which node\n<\/code><\/pre><\/div>\n<p>Update <code>ProgramArguments<\/code> if path differs.<\/p>\n<h3>Load LaunchAgent<\/h3>\n<div><pre><code>launchctl load ~\/Library\/LaunchAgents\/com.openclaw.agent.plist\n<\/code><\/pre><\/div>\n<p><strong>Check if running:<\/strong><\/p>\n<div><pre><code>launchctl list | grep openclaw\n<\/code><\/pre><\/div>\n<p>Should show PID and status.<\/p>\n<p><strong>View logs:<\/strong><\/p>\n<div><pre><code>tail -f ~\/Documents\/openclaw\/logs\/stdout.log\n<\/code><\/pre><\/div>\n<h3>Management Commands<\/h3>\n<div><pre><code>launchctl start com.openclaw.agent   # Start\nlaunchctl stop com.openclaw.agent    # Stop\nlaunchctl unload ~\/Library\/LaunchAgents\/com.openclaw.agent.plist  # Disable auto-start\nlaunchctl load ~\/Library\/LaunchAgents\/com.openclaw.agent.plist    # Enable auto-start\n<\/code><\/pre><\/div>\n<h2 id=\"troubleshooting-arm-specific-issues\">Troubleshooting ARM-Specific Issues<\/h2>\n<h3>&#8220;Cannot find module&#8221; errors<\/h3>\n<p><strong>Symptom:<\/strong><\/p>\n<div><pre><code>Error: Cannot find module '..\/build\/Release\/binding.node'\n<\/code><\/pre><\/div>\n<p><strong>Cause:<\/strong> Native module not compiled for ARM64.<\/p>\n<p><strong>Fix:<\/strong><\/p>\n<div><pre><code>npm rebuild\n<\/code><\/pre><\/div>\n<p>If specific package:<\/p>\n<div><pre><code>npm rebuild [package-name]\n<\/code><\/pre><\/div>\n<h3>Python package errors<\/h3>\n<p><strong>Symptom:<\/strong><\/p>\n<div><pre><code>Error: Command failed: python3 ...\n<\/code><\/pre><\/div>\n<p><strong>Cause:<\/strong> Python script expecting x86 packages.<\/p>\n<p><strong>Fix:<\/strong> Use ARM-compatible Python:<\/p>\n<div><pre><code>brew install python@3.11\n<\/code><\/pre><\/div>\n<p>Then reinstall Python dependencies.<\/p>\n<h3>Chromium\/Puppeteer warnings<\/h3>\n<p><strong>Symptom:<\/strong><\/p>\n<div><pre><code>Warning: Chromium binary is for x86_64, not arm64\n<\/code><\/pre><\/div>\n<p><strong>Cause:<\/strong> Puppeteer downloads Intel Chromium by default.<\/p>\n<p><strong>Fix:<\/strong> Force ARM Chromium:<\/p>\n<div><pre><code>PUPPETEER_SKIP_CHROMIUM_DOWNLOAD=true npm install puppeteer\n<\/code><\/pre><\/div>\n<p>Then set Chromium path to system Chrome:<\/p>\n<div><pre><code>const browser = await puppeteer.launch({\n  executablePath: '\/Applications\/Google Chrome.app\/Contents\/MacOS\/Google Chrome'\n});\n<\/code><\/pre><\/div>\n<h3>Memory issues on base 8GB models<\/h3>\n<p><strong>Symptom:<\/strong> Slow performance, &#8220;out of memory&#8221; errors.<\/p>\n<p><strong>Cause:<\/strong> 8GB shared between system, apps, and GPU.<\/p>\n<p><strong>Fix:<\/strong><\/p>\n<p><strong>Close other apps<\/strong> when running heavy OpenClaw skills.<\/p>\n<p><strong>Disable memory-intensive skills:<\/strong><\/p>\n<div><pre><code>SKILLS_DISABLED=image-gen,video-processing\n<\/code><\/pre><\/div>\n<p><strong>Upgrade to 16GB model<\/strong> if running multiple agents or heavy workloads.<\/p>\n<h2 id=\"docker-on-apple-silicon\">Docker on Apple Silicon<\/h2>\n<p>Docker Desktop for Mac uses virtualization layer. Slight overhead vs native.<\/p>\n<h3>Install Docker Desktop<\/h3>\n<p>Download from docker.com \u2192 Docker Desktop for Mac (Apple Silicon).<\/p>\n<p><strong>Don&#8217;t download Intel version.<\/strong><\/p>\n<h3>Run OpenClaw Container<\/h3>\n<div><pre><code>docker pull openclaw\/openclaw:latest\ndocker run -d --name openclaw \n  -v $(pwd)\/data:\/app\/data \n  -e TELEGRAM_BOT_TOKEN=your-token \n  openclaw\/openclaw:latest\n<\/code><\/pre><\/div>\n<p><strong>Performance:<\/strong> 10-15% slower than native due to VM overhead.<\/p>\n<p><strong>When to use Docker:<\/strong><\/p>\n<ul>\n<li>Testing multiple configurations<\/li>\n<li>Isolation from main system<\/li>\n<li>Deploying same image to cloud<\/li>\n<\/ul>\n<p><strong>When to skip Docker:<\/strong><\/p>\n<ul>\n<li>Maximum performance needed<\/li>\n<li>Running on battery (VM uses more power)<\/li>\n<li>Simple single-instance setup<\/li>\n<\/ul>\n<h2 id=\"the-cost-benefit-analysis\">The Cost-Benefit Analysis<\/h2>\n<h3>Hardware Investment<\/h3>\n<p><strong>Base options:<\/strong><\/p>\n<ul>\n<li>M1 Mac mini (8GB): $599<\/li>\n<li>M2 MacBook Air (8GB): $1,199<\/li>\n<li>M3 MacBook Pro (8GB): $1,599<\/li>\n<\/ul>\n<p><strong>Recommended:<\/strong><\/p>\n<ul>\n<li>M1 Mac mini (16GB): $799 (refurb) &#8211; Best value<\/li>\n<li>M2 MacBook Air (16GB): $1,499 &#8211; Portable option<\/li>\n<li>M3 Pro (18GB): $1,999 &#8211; Power user<\/li>\n<\/ul>\n<h3>Operating Costs<\/h3>\n<p><strong>Electricity:<\/strong><\/p>\n<ul>\n<li>M1 Mac mini idle: 7W = $0.90\/month<\/li>\n<li>M1 Mac mini load: 25W = $3.20\/month<\/li>\n<\/ul>\n<p><strong>Compare to cloud:<\/strong><\/p>\n<ul>\n<li>VPS (2 CPU, 4GB RAM): $10-20\/month<\/li>\n<li>PaioClaw: $15-25\/month<\/li>\n<\/ul>\n<p><strong>Break-even:<\/strong> After 3-4 months, Mac mini is cheaper than cloud (if you use it).<\/p>\n<h3>When Mac Makes Sense<\/h3>\n<p><strong>Choose Mac if:<\/strong><\/p>\n<ul>\n<li>You already own M-series Mac<\/li>\n<li>Want local\/private AI agent<\/li>\n<li>Need offline capability<\/li>\n<li>Development\/learning use case<\/li>\n<li>Long-term (1+ year) usage<\/li>\n<\/ul>\n<p><strong>Choose cloud if:<\/strong><\/p>\n<ul>\n<li>Don&#8217;t own Mac<\/li>\n<li>Need 99.9% uptime<\/li>\n<li>Want zero maintenance<\/li>\n<li>Temporary project (&lt;6 months)<\/li>\n<li>Team\/multi-user deployment<\/li>\n<\/ul>\n<h2 id=\"the-bottom-line\">The Bottom Line<\/h2>\n<p>Apple Silicon Macs are excellent OpenClaw hosts. Native ARM64 Node.js performs 2-3x faster than Intel. Most dependencies &#8220;just work.&#8221; The few that don&#8217;t are fixable with simple rebuilds.<\/p>\n<p><strong>The Homebrew path difference is the biggest gotcha<\/strong> &#8211; remember <code>\/opt\/homebrew\/<\/code> not <code>\/usr\/local\/<\/code>.<\/p>\n<p><strong>Performance hierarchy:<\/strong><\/p>\n<ol>\n<li>Native ARM64 Node + native modules (fastest)<\/li>\n<li>Native Node + few Rosetta modules (still fast)<\/li>\n<li>Node via Rosetta (slow, avoid)<\/li>\n<\/ol>\n<p>If you followed this guide, you now have OpenClaw running natively on Apple Silicon at peak performance.<\/p>\n<p>Whether you run on your Mac or cloud depends on your use case. Mac for personal\/dev. Cloud for production\/team.<\/p>\n<div><\/div>\n<p><strong>Skip the setup entirely?<\/strong> PaioClaw runs OpenClaw in the cloud with zero configuration. No Homebrew path issues, no native module debugging, no battery drain on your laptop. Starts FREE, Smart $15\/month, Genius $25\/month. <a href=\"https:\/\/paioclaw.ai\" target=\"_blank\" rel=\"noopener noreferrer\">Start free \u2192<\/a><\/p><\/div>\n","protected":false},"excerpt":{"rendered":"<p>PaioClaw gives you a private, always-on AI assistant powered by your own API keys. No Docker, no command line \u2014 sign up and it&#8217;s ready in 60 seconds.<\/p>\n","protected":false},"author":0,"featured_media":186,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[2],"tags":[],"class_list":["post-185","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-how-to"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v27.7 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>How to Install OpenClaw on Apple Silicon (M1, M2, M3, M4) - PaioClaw<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/paioclaw.ai\/blog\/openclaw-apple-silicon-setup\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"How to Install OpenClaw on Apple Silicon (M1, M2, M3, M4) - PaioClaw\" \/>\n<meta property=\"og:description\" content=\"PaioClaw gives you a private, always-on AI assistant powered by your own API keys. No Docker, no command line \u2014 sign up and it&#039;s ready in 60 seconds.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/paioclaw.ai\/blog\/openclaw-apple-silicon-setup\/\" \/>\n<meta property=\"og:site_name\" content=\"PaioClaw\" \/>\n<meta property=\"article:publisher\" content=\"https:\/\/www.facebook.com\/paioclaw\/\" \/>\n<meta property=\"article:published_time\" content=\"2026-05-10T00:00:00+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/paioclaw.ai\/blog\/wp-content\/uploads\/2026\/05\/openclaw-apple-silicon-setup.png\" \/>\n\t<meta property=\"og:image:width\" content=\"852\" \/>\n\t<meta property=\"og:image:height\" content=\"341\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/png\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@PaioClaw\" \/>\n<meta name=\"twitter:site\" content=\"@PaioClaw\" \/>\n<meta name=\"twitter:label1\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data1\" content=\"11 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\\\/\\\/paioclaw.ai\\\/blog\\\/openclaw-apple-silicon-setup\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/paioclaw.ai\\\/blog\\\/openclaw-apple-silicon-setup\\\/\"},\"author\":{\"name\":\"\",\"@id\":\"\"},\"headline\":\"How to Install OpenClaw on Apple Silicon (M1, M2, M3, M4)\",\"datePublished\":\"2026-05-10T00:00:00+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/paioclaw.ai\\\/blog\\\/openclaw-apple-silicon-setup\\\/\"},\"wordCount\":1627,\"publisher\":{\"@id\":\"https:\\\/\\\/paioclaw.ai\\\/blog\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/paioclaw.ai\\\/blog\\\/openclaw-apple-silicon-setup\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/paioclaw.ai\\\/blog\\\/wp-content\\\/uploads\\\/2026\\\/05\\\/openclaw-apple-silicon-setup.png\",\"articleSection\":[\"How to\"],\"inLanguage\":\"en-US\"},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/paioclaw.ai\\\/blog\\\/openclaw-apple-silicon-setup\\\/\",\"url\":\"https:\\\/\\\/paioclaw.ai\\\/blog\\\/openclaw-apple-silicon-setup\\\/\",\"name\":\"How to Install OpenClaw on Apple Silicon (M1, M2, M3, M4) - PaioClaw\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/paioclaw.ai\\\/blog\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/paioclaw.ai\\\/blog\\\/openclaw-apple-silicon-setup\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/paioclaw.ai\\\/blog\\\/openclaw-apple-silicon-setup\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/paioclaw.ai\\\/blog\\\/wp-content\\\/uploads\\\/2026\\\/05\\\/openclaw-apple-silicon-setup.png\",\"datePublished\":\"2026-05-10T00:00:00+00:00\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/paioclaw.ai\\\/blog\\\/openclaw-apple-silicon-setup\\\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/paioclaw.ai\\\/blog\\\/openclaw-apple-silicon-setup\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/paioclaw.ai\\\/blog\\\/openclaw-apple-silicon-setup\\\/#primaryimage\",\"url\":\"https:\\\/\\\/paioclaw.ai\\\/blog\\\/wp-content\\\/uploads\\\/2026\\\/05\\\/openclaw-apple-silicon-setup.png\",\"contentUrl\":\"https:\\\/\\\/paioclaw.ai\\\/blog\\\/wp-content\\\/uploads\\\/2026\\\/05\\\/openclaw-apple-silicon-setup.png\",\"width\":852,\"height\":341},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/paioclaw.ai\\\/blog\\\/openclaw-apple-silicon-setup\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/paioclaw.ai\\\/blog\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"How to Install OpenClaw on Apple Silicon (M1, M2, M3, M4)\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/paioclaw.ai\\\/blog\\\/#website\",\"url\":\"https:\\\/\\\/paioclaw.ai\\\/blog\\\/\",\"name\":\"PAIO Blog \u2014 Guides, tips, and updates\",\"description\":\"\",\"publisher\":{\"@id\":\"https:\\\/\\\/paioclaw.ai\\\/blog\\\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/paioclaw.ai\\\/blog\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/paioclaw.ai\\\/blog\\\/#organization\",\"name\":\"PAIO\",\"url\":\"https:\\\/\\\/paioclaw.ai\\\/blog\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/paioclaw.ai\\\/blog\\\/#\\\/schema\\\/logo\\\/image\\\/\",\"url\":\"https:\\\/\\\/local.paioclawblog.com\\\/wp-content\\\/uploads\\\/2026\\\/05\\\/paioclaw_logo.webp\",\"contentUrl\":\"https:\\\/\\\/local.paioclawblog.com\\\/wp-content\\\/uploads\\\/2026\\\/05\\\/paioclaw_logo.webp\",\"width\":128,\"height\":128,\"caption\":\"PAIO\"},\"image\":{\"@id\":\"https:\\\/\\\/paioclaw.ai\\\/blog\\\/#\\\/schema\\\/logo\\\/image\\\/\"},\"sameAs\":[\"https:\\\/\\\/www.facebook.com\\\/paioclaw\\\/\",\"https:\\\/\\\/x.com\\\/PaioClaw\",\"https:\\\/\\\/www.instagram.com\\\/paioclaw\\\/\",\"https:\\\/\\\/www.linkedin.com\\\/company\\\/paioclaw\",\"https:\\\/\\\/www.youtube.com\\\/@PaioClaw\"]}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"How to Install OpenClaw on Apple Silicon (M1, M2, M3, M4) - PaioClaw","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/paioclaw.ai\/blog\/openclaw-apple-silicon-setup\/","og_locale":"en_US","og_type":"article","og_title":"How to Install OpenClaw on Apple Silicon (M1, M2, M3, M4) - PaioClaw","og_description":"PaioClaw gives you a private, always-on AI assistant powered by your own API keys. No Docker, no command line \u2014 sign up and it's ready in 60 seconds.","og_url":"https:\/\/paioclaw.ai\/blog\/openclaw-apple-silicon-setup\/","og_site_name":"PaioClaw","article_publisher":"https:\/\/www.facebook.com\/paioclaw\/","article_published_time":"2026-05-10T00:00:00+00:00","og_image":[{"width":852,"height":341,"url":"https:\/\/paioclaw.ai\/blog\/wp-content\/uploads\/2026\/05\/openclaw-apple-silicon-setup.png","type":"image\/png"}],"twitter_card":"summary_large_image","twitter_creator":"@PaioClaw","twitter_site":"@PaioClaw","twitter_misc":{"Est. reading time":"11 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/paioclaw.ai\/blog\/openclaw-apple-silicon-setup\/#article","isPartOf":{"@id":"https:\/\/paioclaw.ai\/blog\/openclaw-apple-silicon-setup\/"},"author":{"name":"","@id":""},"headline":"How to Install OpenClaw on Apple Silicon (M1, M2, M3, M4)","datePublished":"2026-05-10T00:00:00+00:00","mainEntityOfPage":{"@id":"https:\/\/paioclaw.ai\/blog\/openclaw-apple-silicon-setup\/"},"wordCount":1627,"publisher":{"@id":"https:\/\/paioclaw.ai\/blog\/#organization"},"image":{"@id":"https:\/\/paioclaw.ai\/blog\/openclaw-apple-silicon-setup\/#primaryimage"},"thumbnailUrl":"https:\/\/paioclaw.ai\/blog\/wp-content\/uploads\/2026\/05\/openclaw-apple-silicon-setup.png","articleSection":["How to"],"inLanguage":"en-US"},{"@type":"WebPage","@id":"https:\/\/paioclaw.ai\/blog\/openclaw-apple-silicon-setup\/","url":"https:\/\/paioclaw.ai\/blog\/openclaw-apple-silicon-setup\/","name":"How to Install OpenClaw on Apple Silicon (M1, M2, M3, M4) - PaioClaw","isPartOf":{"@id":"https:\/\/paioclaw.ai\/blog\/#website"},"primaryImageOfPage":{"@id":"https:\/\/paioclaw.ai\/blog\/openclaw-apple-silicon-setup\/#primaryimage"},"image":{"@id":"https:\/\/paioclaw.ai\/blog\/openclaw-apple-silicon-setup\/#primaryimage"},"thumbnailUrl":"https:\/\/paioclaw.ai\/blog\/wp-content\/uploads\/2026\/05\/openclaw-apple-silicon-setup.png","datePublished":"2026-05-10T00:00:00+00:00","breadcrumb":{"@id":"https:\/\/paioclaw.ai\/blog\/openclaw-apple-silicon-setup\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/paioclaw.ai\/blog\/openclaw-apple-silicon-setup\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/paioclaw.ai\/blog\/openclaw-apple-silicon-setup\/#primaryimage","url":"https:\/\/paioclaw.ai\/blog\/wp-content\/uploads\/2026\/05\/openclaw-apple-silicon-setup.png","contentUrl":"https:\/\/paioclaw.ai\/blog\/wp-content\/uploads\/2026\/05\/openclaw-apple-silicon-setup.png","width":852,"height":341},{"@type":"BreadcrumbList","@id":"https:\/\/paioclaw.ai\/blog\/openclaw-apple-silicon-setup\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/paioclaw.ai\/blog\/"},{"@type":"ListItem","position":2,"name":"How to Install OpenClaw on Apple Silicon (M1, M2, M3, M4)"}]},{"@type":"WebSite","@id":"https:\/\/paioclaw.ai\/blog\/#website","url":"https:\/\/paioclaw.ai\/blog\/","name":"PAIO Blog \u2014 Guides, tips, and updates","description":"","publisher":{"@id":"https:\/\/paioclaw.ai\/blog\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/paioclaw.ai\/blog\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Organization","@id":"https:\/\/paioclaw.ai\/blog\/#organization","name":"PAIO","url":"https:\/\/paioclaw.ai\/blog\/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/paioclaw.ai\/blog\/#\/schema\/logo\/image\/","url":"https:\/\/local.paioclawblog.com\/wp-content\/uploads\/2026\/05\/paioclaw_logo.webp","contentUrl":"https:\/\/local.paioclawblog.com\/wp-content\/uploads\/2026\/05\/paioclaw_logo.webp","width":128,"height":128,"caption":"PAIO"},"image":{"@id":"https:\/\/paioclaw.ai\/blog\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/www.facebook.com\/paioclaw\/","https:\/\/x.com\/PaioClaw","https:\/\/www.instagram.com\/paioclaw\/","https:\/\/www.linkedin.com\/company\/paioclaw","https:\/\/www.youtube.com\/@PaioClaw"]}]}},"_links":{"self":[{"href":"https:\/\/paioclaw.ai\/blog\/wp-json\/wp\/v2\/posts\/185","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/paioclaw.ai\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/paioclaw.ai\/blog\/wp-json\/wp\/v2\/types\/post"}],"replies":[{"embeddable":true,"href":"https:\/\/paioclaw.ai\/blog\/wp-json\/wp\/v2\/comments?post=185"}],"version-history":[{"count":0,"href":"https:\/\/paioclaw.ai\/blog\/wp-json\/wp\/v2\/posts\/185\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/paioclaw.ai\/blog\/wp-json\/wp\/v2\/media\/186"}],"wp:attachment":[{"href":"https:\/\/paioclaw.ai\/blog\/wp-json\/wp\/v2\/media?parent=185"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/paioclaw.ai\/blog\/wp-json\/wp\/v2\/categories?post=185"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/paioclaw.ai\/blog\/wp-json\/wp\/v2\/tags?post=185"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}