{"id":2717,"date":"2026-04-19T09:38:03","date_gmt":"2026-04-19T09:38:03","guid":{"rendered":"https:\/\/primetoolhub.com\/?page_id=2717"},"modified":"2026-11-10T12:57:37","modified_gmt":"2026-11-10T12:57:37","slug":"browser-ai-models-directory","status":"publish","type":"page","link":"https:\/\/schoolict.net\/tools\/browser-ai-models-directory\/","title":{"rendered":"Ultimate Browser AI Models Directory 2026 | Offline WebGPU WASM Models"},"content":{"rendered":"<div class=\"pth-hero-section\">\n<div class=\"pth-hero-content\">\n<h2>Ultimate Browser AI Models Directory <\/h2>\n<p>Ultimate 2026 browser ai models directory with 85+ models, clear \ud83d\udfe2 offline browser capable highlighting, and new \u201cComputer Languages \/ Integration\u201d column showing exactly which languages &#038; frameworks (JavaScript, WebGPU, WASM, ONNX, TF.js etc.) can handle &#038; connect each model. Perfect reference for PrimeToolHub offline web tools.<\/p>\n<div id=\"pth-toc-placeholder\"><\/div>\n<\/p>\n<\/div>\n<div class=\"pth-hero-image\">\n    <img data-no-lazy=\"1\"\n         src=\"https:\/\/schoolict.net\/tools\/wp-content\/uploads\/2026\/04\/browser-ai-models-directory-800x447.jpeg\"\n         width=\"800\"\n         height=\"447\"\n         alt=\"browser ai models directory\"\n         fetchpriority=\"high\"\n         loading=\"eager\"\n         decoding=\"async\"\n         style=\"width:100%; height:auto; display:block;\">\n  <\/div>\n<\/div>\n\n\n\n<div class=\"wp-block-rank-math-toc-block\" id=\"rank-math-toc\"><h2>Table of Contents<\/h2><nav><ul><li><a href=\"#\ud83d\udd34-how-to-read-the-directory-without-getting-lost\">\ud83d\udd34\u00a0How to read the directory without getting lost<\/a><\/li><li><a href=\"#\ud83d\udfe1-a-few-real-searches-and-what-comes-back\">\ud83d\udfe1\u00a0A few real searches, and what comes back<\/a><\/li><li><a href=\"#\ud83d\udfe2-what-this-directory-will-not-do-for-you\">\ud83d\udfe2\u00a0What this directory will not do for you<\/a><\/li><\/ul><\/nav><\/div>\n\n\n<style>\r\n\/* ==========================================================================\r\n   PRIME TOOL HUB - BROWSER AI MODELS DIRECTORY (TOOL 52)\r\n   Engine: 100% Cache-Proof, Advanced Filters, 80+ Curated Models\r\n   ========================================================================== *\/\r\n#pth-ai-models-dir-wrapper { font-family: 'Inter', system-ui, sans-serif; max-width: 1500px; margin: 0 auto; box-sizing: border-box; background: #f8fafc; padding: 20px; border-radius: 16px; color: #0f172a; position: relative; }\r\n#pth-ai-models-dir-wrapper * { box-sizing: border-box; min-width: 0; }\r\n\r\n\/* Base Layout & Top Bar *\/\r\n#pth-ai-models-dir-wrapper .pth-top-bar { background: #ffffff; border: 1px solid #e2e8f0; padding: 20px 25px; border-radius: 12px; display: flex; justify-content: space-between; align-items: center; margin-bottom: 25px; box-shadow: 0 4px 6px -1px rgba(0,0,0,0.05); flex-wrap: wrap; gap: 15px;}\r\n#pth-ai-models-dir-wrapper .pth-tool-title { font-size: 1.6rem; font-weight: 800; color: #1e293b; margin: 0; display: flex; align-items: center; gap: 10px;}\r\n#pth-ai-models-dir-wrapper .pth-subtitle { font-size: 0.9rem; color: #64748b; font-weight: 600; margin-top: 4px; display: block;}\r\n#pth-ai-models-dir-wrapper .pth-flex-layout { display: grid; grid-template-columns: 1.8fr 1fr; gap: 25px; align-items: start; transition: 0.3s ease; }\r\n#pth-ai-models-dir-wrapper .pth-flex-layout.pth-full-width { grid-template-columns: 1fr; }\r\n@media (max-width: 1024px) { #pth-ai-models-dir-wrapper .pth-flex-layout { grid-template-columns: 1fr; } }\r\n\r\n\/* Cards & Sections *\/\r\n#pth-ai-models-dir-wrapper .pth-card { background: #ffffff; border: 1px solid #e2e8f0; border-radius: 12px; padding: 25px; margin-bottom: 25px; box-shadow: 0 4px 6px -1px rgba(0,0,0,0.02); }\r\n#pth-ai-models-dir-wrapper .pth-card-title { font-size: 1.15rem; font-weight: 800; color: #1e293b; margin: 0 0 15px 0; display: flex; align-items: center; gap: 8px; padding-bottom: 15px; border-bottom: 2px solid #e2e8f0;}\r\n\r\n\/* Filters Area *\/\r\n#pth-ai-models-dir-wrapper .pth-filters-wrap { display: flex; flex-wrap: wrap; gap: 12px; margin-bottom: 20px; align-items: center; background: #f1f5f9; padding: 15px; border-radius: 10px; border: 1px solid #cbd5e1;}\r\n#pth-ai-models-dir-wrapper .pth-search-input { flex: 1; min-width: 220px; padding: 10px 15px; border: 1px solid #cbd5e1; border-radius: 8px; font-size: 0.95rem; outline: none; transition: 0.2s; font-family: 'Inter', sans-serif;}\r\n#pth-ai-models-dir-wrapper .pth-search-input:focus { border-color: #3b82f6; box-shadow: 0 0 0 3px rgba(59, 130, 246, 0.1); }\r\n#pth-ai-models-dir-wrapper .pth-select-filter { padding: 10px 12px; border: 1px solid #cbd5e1; border-radius: 8px; font-size: 0.85rem; background: #fff; cursor: pointer; outline: none; font-weight: 600; color: #334155;}\r\n#pth-ai-models-dir-wrapper .pth-select-filter:focus { border-color: #3b82f6; }\r\n\r\n\/* Table Styles *\/\r\n#pth-ai-models-dir-wrapper .pth-table-container { width: 100%; overflow-x: auto; border-radius: 8px; border: 1px solid #e2e8f0; }\r\n#pth-ai-models-dir-wrapper .pth-models-table { width: 100%; border-collapse: collapse; text-align: left; background: #fff; min-width: 900px;}\r\n#pth-ai-models-dir-wrapper .pth-models-table th { background: #f8fafc; padding: 14px 12px; font-weight: 700; font-size: 0.8rem; color: #475569; border-bottom: 2px solid #e2e8f0; text-transform: uppercase; letter-spacing: 0.5px;}\r\n#pth-ai-models-dir-wrapper .pth-models-table th.sortable { cursor: pointer; transition: background 0.2s; }\r\n#pth-ai-models-dir-wrapper .pth-models-table th.sortable:hover { background: #f1f5f9; }\r\n#pth-ai-models-dir-wrapper .pth-models-table td { padding: 13px 12px; border-bottom: 1px solid #e2e8f0; font-size: 0.88rem; color: #1e293b; vertical-align: top;}\r\n#pth-ai-models-dir-wrapper .pth-models-table tbody tr:hover { background: #f8fafc; }\r\n#pth-ai-models-dir-wrapper .pth-models-table tbody tr:last-child td { border-bottom: none; }\r\n\r\n\/* Tags & Badges *\/\r\n#pth-ai-models-dir-wrapper .pth-tag { display: inline-block; padding: 3px 9px; border-radius: 20px; font-size: 0.72rem; font-weight: 700; margin-bottom: 4px;}\r\n#pth-ai-models-dir-wrapper .pth-tag-local { background: #dcfce7; color: #047857; border: 1px solid #a7f3d0; }\r\n#pth-ai-models-dir-wrapper .pth-tag-cloud { background: #fee2e2; color: #b91c1c; border: 1px solid #fecaca; }\r\n#pth-ai-models-dir-wrapper .pth-tag-cat { background: #f1f5f9; color: #475569; border: 1px solid #cbd5e1; }\r\n#pth-ai-models-dir-wrapper .pth-count-badge { display: inline-block; background: #3b82f6; color: white; border-radius: 20px; padding: 2px 10px; font-size: 0.8rem; font-weight: 800; margin-left: 8px; }\r\n\r\n\/* Language Badges *\/\r\n#pth-ai-models-dir-wrapper .pth-langs-section { margin-bottom: 8px; }\r\n#pth-ai-models-dir-wrapper .pth-langs-label { font-size: 0.7rem; font-weight: 700; color: #475569; text-transform: uppercase; letter-spacing: 0.4px; display: block; margin-bottom: 4px; }\r\n#pth-ai-models-dir-wrapper .pth-lang-wrap { display: flex; flex-wrap: wrap; gap: 3px; }\r\n#pth-ai-models-dir-wrapper .pth-lb { display: inline-block; padding: 2px 7px; border-radius: 10px; font-size: 0.68rem; font-weight: 700; white-space: nowrap; }\r\n#pth-ai-models-dir-wrapper .lb-py  { background: #dbeafe; color: #1d4ed8; }\r\n#pth-ai-models-dir-wrapper .lb-js  { background: #fef9c3; color: #854d0e; }\r\n#pth-ai-models-dir-wrapper .lb-ts  { background: #e0f2fe; color: #0369a1; }\r\n#pth-ai-models-dir-wrapper .lb-java{ background: #fff7ed; color: #c2410c; }\r\n#pth-ai-models-dir-wrapper .lb-cpp { background: #ede9fe; color: #5b21b6; }\r\n#pth-ai-models-dir-wrapper .lb-cs  { background: #f3e8ff; color: #7e22ce; }\r\n#pth-ai-models-dir-wrapper .lb-go  { background: #e0fdf4; color: #065f46; }\r\n#pth-ai-models-dir-wrapper .lb-rs  { background: #fef2f2; color: #991b1b; }\r\n#pth-ai-models-dir-wrapper .lb-php { background: #f5f3ff; color: #4c1d95; }\r\n#pth-ai-models-dir-wrapper .lb-rb  { background: #fce7f3; color: #9d174d; }\r\n#pth-ai-models-dir-wrapper .lb-sw  { background: #fff7ed; color: #9a3412; }\r\n#pth-ai-models-dir-wrapper .lb-sql { background: #ecfdf5; color: #065f46; }\r\n#pth-ai-models-dir-wrapper .lb-sh  { background: #f1f5f9; color: #334155; }\r\n#pth-ai-models-dir-wrapper .lb-r   { background: #eff6ff; color: #1e40af; }\r\n#pth-ai-models-dir-wrapper .lb-kt  { background: #fdf4ff; color: #86198f; }\r\n#pth-ai-models-dir-wrapper .lb-na  { background: #f8fafc; color: #475569; font-style: italic; }\r\n#pth-ai-models-dir-wrapper .lb-curl{ background: #f0fdf4; color: #166534; }\r\n#pth-ai-models-dir-wrapper .lb-lua { background: #fef3c7; color: #92400e; }\r\n#pth-ai-models-dir-wrapper .pth-lang-divider { border: none; border-top: 1px dashed #e2e8f0; margin: 7px 0; }\r\n\r\n\/* Empty State *\/\r\n#pth-ai-models-dir-wrapper .pth-empty-state { padding: 40px 20px; text-align: center; color: #64748b; display: none; }\r\n#pth-ai-models-dir-wrapper .pth-empty-state h4 { margin: 0 0 10px 0; color: #1e293b; font-size: 1.1rem; }\r\n\r\n\/* Buttons *\/\r\n#pth-ai-models-dir-wrapper .pth-btn { padding: 10px 20px; border-radius: 8px; font-weight: 700; font-size: 0.9rem; cursor: pointer; transition: 0.2s; display: inline-flex; align-items: center; justify-content: center; gap: 8px; border: none; font-family: 'Inter', sans-serif;}\r\n#pth-ai-models-dir-wrapper .pth-btn-primary { background: #3b82f6; color: white; box-shadow: 0 4px 10px rgba(59,130,246,0.25); }\r\n#pth-ai-models-dir-wrapper .pth-btn-primary:hover { background: #2563eb; transform: translateY(-1px); }\r\n#pth-ai-models-dir-wrapper .pth-btn-success { background: #10b981; color: white; box-shadow: 0 4px 10px rgba(16,185,129,0.25); }\r\n#pth-ai-models-dir-wrapper .pth-btn-success:hover { background: #059669; transform: translateY(-1px); }\r\n#pth-ai-models-dir-wrapper .pth-btn-secondary { background: #e2e8f0; color: #475569; }\r\n#pth-ai-models-dir-wrapper .pth-btn-secondary:hover { background: #cbd5e1; color: #0f172a;}\r\n#pth-ai-models-dir-wrapper .pth-btn-dark { background: #0f172a; color: white; }\r\n#pth-ai-models-dir-wrapper .pth-btn-dark:hover { background: #1e293b; }\r\n#pth-ai-models-dir-wrapper .pth-btn-small { padding: 6px 12px; font-size: 0.8rem; border-radius: 6px; }\r\n\r\n#pth-ai-models-dir-wrapper .pth-sample-btn { background: #ecfdf5; color: #10b981; border: 1px solid #6ee7b7; font-size: 0.8rem; padding: 6px 12px; border-radius: 6px; font-weight: 700; cursor: pointer; transition: 0.2s; display: inline-flex; align-items: center; gap: 6px; }\r\n#pth-ai-models-dir-wrapper .pth-sample-btn:hover { background: #d1fae5; }\r\n\r\n\/* Sidebar Elements *\/\r\n#pth-ai-models-dir-wrapper .pth-sidebar-col { position: sticky; top: 20px; height: fit-content; }\r\n#pth-ai-models-dir-wrapper .pth-sidebar-list { padding-left: 20px; font-size: 0.9rem; color: #475569; line-height: 1.6; margin: 0;}\r\n#pth-ai-models-dir-wrapper .pth-sidebar-list li { margin-bottom: 8px; }\r\n#pth-ai-models-dir-wrapper .pth-lang-legend { display: flex; flex-wrap: wrap; gap: 6px; margin-top: 10px; }\r\n\r\n\/* Toast *\/\r\n#pth-ai-models-dir-wrapper .pth-toast { position: fixed; bottom: 20px; right: 20px; background: #1e293b; color: white; padding: 12px 24px; border-radius: 8px; font-weight: 700; font-size: 0.9rem; box-shadow: 0 10px 30px rgba(0,0,0,0.3); z-index: 9999; display: none; animation: pthSlideIn 0.3s ease; }\r\n#pth-ai-models-dir-wrapper .pth-toast.show { display: flex; align-items:center; gap:8px; }\r\n@keyframes pthSlideIn { from { opacity: 0; transform: translateY(20px); } to { opacity: 1; transform: translateY(0); } }\r\n<\/style>\r\n\r\n<div id=\"pth-ai-models-dir-wrapper\">\r\n    <div id=\"pth-dir-toast\" class=\"pth-toast\">\u2705 Action Successful!<\/div>\r\n\r\n    <div class=\"pth-top-bar\">\r\n        <div>\r\n            <h2 class=\"pth-tool-title\">\ud83e\udde0 Ultimate AI Models Directory (2026)<\/h2>\r\n            <span class=\"pth-subtitle\">Curated models \u2022 \ud83d\udfe2 Offline Browser vs \ud83d\udd34 Cloud API \u2022 Code Support Analysis<\/span>\r\n        <\/div>\r\n        <div style=\"display:flex; align-items:center; gap:10px; flex-wrap:wrap;\">\r\n            <button class=\"pth-sample-btn\" id=\"pth-load-sample-btn\">\ud83d\udca1 Try: Code LLMs (Python)<\/button>\r\n            <button class=\"pth-btn pth-btn-secondary pth-btn-small\" id=\"pth-fw-btn-ai-dir\">\u26f6 Full Width<\/button>\r\n        <\/div>\r\n    <\/div>\r\n\r\n    <div class=\"pth-flex-layout\" id=\"pth-ai-models-dir-layout\">\r\n        \r\n        <div class=\"pth-main-col\">\r\n            \r\n            <div class=\"pth-card\" style=\"padding:15px 25px;\">\r\n                <div class=\"pth-filters-wrap\" style=\"margin-bottom:15px;\">\r\n                    <input type=\"text\" id=\"pth-dir-search\" class=\"pth-search-input\" placeholder=\"\ud83d\udd0d Search models, tech, languages, use cases...\">\r\n                    \r\n                    <select id=\"pth-filter-cat\" class=\"pth-select-filter\">\r\n                        <option value=\"\">All Categories<\/option>\r\n                        <option value=\"Text \/ LLM\">\ud83d\udcdd Text \/ LLM<\/option>\r\n                        <option value=\"Image\">\ud83d\uddbc\ufe0f Image<\/option>\r\n                        <option value=\"Video\">\ud83c\udfac Video<\/option>\r\n                        <option value=\"Audio \/ Speech\">\ud83d\udd0a Audio \/ Speech<\/option>\r\n                        <option value=\"Vision \/ Utils\">\ud83d\udc41\ufe0f Vision \/ Utilities<\/option>\r\n                        <option value=\"Embedding\">\ud83d\udd0e Embedding \/ RAG<\/option>\r\n                    <\/select>\r\n\r\n                    <select id=\"pth-filter-type\" class=\"pth-select-filter\">\r\n                        <option value=\"\">All Deployments<\/option>\r\n                        <option value=\"Local \/ Browser\">\ud83d\udfe2 Offline Browser Capable<\/option>\r\n                        <option value=\"Cloud API\">\ud83d\udd34 Cloud Only<\/option>\r\n                    <\/select>\r\n\r\n                    <select id=\"pth-filter-lang\" class=\"pth-select-filter\">\r\n                        <option value=\"\">All Languages<\/option>\r\n                        <option value=\"Python\">\ud83d\udc0d Python<\/option>\r\n                        <option value=\"JavaScript\">\ud83c\udf10 JavaScript<\/option>\r\n                        <option value=\"TypeScript\">\ud83d\udd37 TypeScript<\/option>\r\n                        <option value=\"Java\">\u2615 Java<\/option>\r\n                        <option value=\"C++\">\u2699\ufe0f C++<\/option>\r\n                        <option value=\"C#\">\ud83d\udd37 C#<\/option>\r\n                        <option value=\"Go\">\ud83d\udc39 Go<\/option>\r\n                        <option value=\"Rust\">\ud83e\udd80 Rust<\/option>\r\n                        <option value=\"PHP\">\ud83d\udc18 PHP<\/option>\r\n                        <option value=\"Ruby\">\ud83d\udc8e Ruby<\/option>\r\n                        <option value=\"Swift\">\ud83c\udf4e Swift<\/option>\r\n                        <option value=\"SQL\">\ud83d\uddc4\ufe0f SQL<\/option>\r\n                        <option value=\"Bash\">\ud83d\udcbb Bash \/ Shell<\/option>\r\n                        <option value=\"Kotlin\">\ud83c\udfaf Kotlin<\/option>\r\n                        <option value=\"R\">\ud83d\udcca R<\/option>\r\n                    <\/select>\r\n\r\n                    <button class=\"pth-btn pth-btn-secondary pth-btn-small\" id=\"pth-reset-filters-btn\">\ud83d\udd04 Reset<\/button>\r\n                <\/div>\r\n\r\n                <div style=\"display:flex; align-items:center; justify-content:space-between; margin-bottom:12px;\">\r\n                    <span style=\"font-size:0.85rem; color:#64748b; font-weight:600;\">\r\n                        Showing <span id=\"pth-result-count\" class=\"pth-count-badge\">0<\/span> models\r\n                    <\/span>\r\n                <\/div>\r\n\r\n                <div class=\"pth-table-container\">\r\n                    <table class=\"pth-models-table\" id=\"pth-models-table\">\r\n                        <thead>\r\n                            <tr>\r\n                                <th class=\"sortable\" data-sort=\"name\" style=\"min-width:160px;\">Model Name \u2195<\/th>\r\n                                <th class=\"sortable\" data-sort=\"cat\" style=\"min-width:110px;\">Category \u2195<\/th>\r\n                                <th style=\"min-width:180px;\">Deployment & Tech<\/th>\r\n                                <th style=\"min-width:200px;\">\ud83d\udda5\ufe0f Handles Code In & \ud83d\udd17 Connect SDK<\/th>\r\n                                <th style=\"min-width:180px;\">Primary Use Case<\/th>\r\n                            <\/tr>\r\n                        <\/thead>\r\n                        <tbody id=\"pth-table-body\">\r\n                            <\/tbody>\r\n                    <\/table>\r\n                <\/div>\r\n\r\n                <div class=\"pth-empty-state\" id=\"pth-empty-state\">\r\n                    <h4>\ud83d\udfe8 No models found<\/h4>\r\n                    <p>Try adjusting your search terms or filters above.<\/p>\r\n                <\/div>\r\n            <\/div>\r\n\r\n        <\/div>\r\n\r\n        <div class=\"pth-sidebar-col\" id=\"pth-ai-models-dir-sidebar\">\r\n            \r\n            <div class=\"pth-card\" style=\"background:#f0fdf4; border-color:#6ee7b7; padding:20px;\">\r\n                <h3 class=\"pth-card-title\" style=\"color:#065f46; border-color:#a7f3d0;\">\u26a1 Quick Actions<\/h3>\r\n                <div style=\"display:flex; flex-direction:column; gap:10px;\">\r\n                    <button class=\"pth-btn pth-btn-success\" id=\"pth-export-csv-btn\">\ud83d\udce5 Export to CSV<\/button>\r\n                    <button class=\"pth-btn pth-btn-secondary\" id=\"pth-print-btn\">\ud83d\udda8\ufe0f Print Table<\/button>\r\n                    <button class=\"pth-btn pth-btn-dark\" id=\"pth-copy-json-btn\">\ud83d\udccb Copy JSON Data<\/button>\r\n                <\/div>\r\n            <\/div>\r\n\r\n            <div class=\"pth-card\" style=\"padding:20px;\">\r\n                <h3 class=\"pth-card-title\" style=\"font-size:1.1rem;\">\ud83c\udff7\ufe0f Language Badge Legend<\/h3>\r\n                <div class=\"pth-lang-legend\">\r\n                    <span class=\"pth-lb lb-py\">Python<\/span>\r\n                    <span class=\"pth-lb lb-js\">JavaScript<\/span>\r\n                    <span class=\"pth-lb lb-ts\">TypeScript<\/span>\r\n                    <span class=\"pth-lb lb-java\">Java<\/span>\r\n                    <span class=\"pth-lb lb-cpp\">C++<\/span>\r\n                    <span class=\"pth-lb lb-cs\">C#<\/span>\r\n                    <span class=\"pth-lb lb-go\">Go<\/span>\r\n                    <span class=\"pth-lb lb-rs\">Rust<\/span>\r\n                    <span class=\"pth-lb lb-php\">PHP<\/span>\r\n                    <span class=\"pth-lb lb-rb\">Ruby<\/span>\r\n                    <span class=\"pth-lb lb-sw\">Swift<\/span>\r\n                    <span class=\"pth-lb lb-sql\">SQL<\/span>\r\n                    <span class=\"pth-lb lb-sh\">Bash<\/span>\r\n                    <span class=\"pth-lb lb-r\">R<\/span>\r\n                    <span class=\"pth-lb lb-kt\">Kotlin<\/span>\r\n                    <span class=\"pth-lb lb-curl\">curl<\/span>\r\n                <\/div>\r\n                <p style=\"font-size:0.78rem; color:#64748b; margin-top:12px; margin-bottom:0;\">\r\n                    <strong>\ud83d\udda5\ufe0f Handles:<\/strong> Code languages the model understands & generates.<br>\r\n                    <strong>\ud83d\udd17 SDK:<\/strong> Languages to connect\/integrate the model into your app.\r\n                <\/p>\r\n            <\/div>\r\n\r\n            <div class=\"pth-card\" style=\"padding:20px;\">\r\n                <h3 class=\"pth-card-title\" style=\"font-size:1.1rem;\">\ud83d\udcbb Why Offline Browser Models?<\/h3>\r\n                <p style=\"font-size:0.87rem; color:#475569; margin-bottom:12px;\">Models marked <strong>\ud83d\udfe2 Offline Browser Capable<\/strong> run 100% client-side (IndexedDB + WebGPU\/WASM\/ONNX). They offer:<\/p>\r\n                <ul class=\"pth-sidebar-list\">\r\n                    <li><strong>100% Privacy:<\/strong> Data never leaves the device<\/li>\r\n                    <li><strong>Zero Server Cost:<\/strong> Runs on user's GPU\/CPU<\/li>\r\n                    <li><strong>Full Offline:<\/strong> Works without internet after first load<\/li>\r\n                    <li><strong>SDK Support:<\/strong> JS, Python, WASM, ONNX Runtime<\/li>\r\n                <\/ul>\r\n            <\/div>\r\n\r\n            <div class=\"pth-card\" style=\"padding:20px;\">\r\n                <h3 class=\"pth-card-title\" style=\"font-size:1.1rem;\">\ud83d\udcca Quick Filters<\/h3>\r\n                <div style=\"display:flex; flex-direction:column; gap:8px;\">\r\n                    <button class=\"pth-sample-btn pth-quick-filter-btn\" data-cat=\"Text \/ LLM\" data-type=\"Local \/ Browser\" data-lang=\"Python\">\ud83d\udc0d Python-ready LLMs (Local)<\/button>\r\n                    <button class=\"pth-sample-btn pth-quick-filter-btn\" data-cat=\"Text \/ LLM\" data-type=\"Cloud API\" data-lang=\"TypeScript\">\ud83d\udd37 TS-SDK Cloud LLMs<\/button>\r\n                    <button class=\"pth-sample-btn pth-quick-filter-btn\" data-cat=\"Embedding\" data-type=\"\" data-lang=\"JavaScript\">\ud83d\udd0e JS Embedding Models<\/button>\r\n                    <button class=\"pth-sample-btn pth-quick-filter-btn\" data-cat=\"Vision \/ Utils\" data-type=\"Local \/ Browser\" data-lang=\"\">\ud83d\udc41\ufe0f Offline Vision Tools<\/button>\r\n                    <button class=\"pth-sample-btn pth-quick-filter-btn\" data-cat=\"Audio \/ Speech\" data-type=\"Local \/ Browser\" data-lang=\"\">\ud83d\udd0a Offline Audio Models<\/button>\r\n                <\/div>\r\n            <\/div>\r\n\r\n        <\/div>\r\n\r\n    <\/div>\r\n<\/div>\r\n\r\n<script data-no-optimize=\"1\" data-no-minify=\"1\" data-cfasync=\"false\">\r\n(function() {\r\n    window.pthAiModelsDirApp = {\r\n        initialized: false,\r\n        currentSort: { col: 'name', dir: 1 },\r\n        \r\n        LB: {\r\n            \"Python\":     { cls:\"lb-py\",   lbl:\"Python\" },\r\n            \"JavaScript\": { cls:\"lb-js\",   lbl:\"JS\" },\r\n            \"TypeScript\": { cls:\"lb-ts\",   lbl:\"TS\" },\r\n            \"Java\":       { cls:\"lb-java\", lbl:\"Java\" },\r\n            \"C++\":        { cls:\"lb-cpp\",  lbl:\"C++\" },\r\n            \"C#\":         { cls:\"lb-cs\",   lbl:\"C#\" },\r\n            \"Go\":         { cls:\"lb-go\",   lbl:\"Go\" },\r\n            \"Rust\":       { cls:\"lb-rs\",   lbl:\"Rust\" },\r\n            \"PHP\":        { cls:\"lb-php\",  lbl:\"PHP\" },\r\n            \"Ruby\":       { cls:\"lb-rb\",   lbl:\"Ruby\" },\r\n            \"Swift\":      { cls:\"lb-sw\",   lbl:\"Swift\" },\r\n            \"SQL\":        { cls:\"lb-sql\",  lbl:\"SQL\" },\r\n            \"Bash\":       { cls:\"lb-sh\",   lbl:\"Bash\" },\r\n            \"R\":          { cls:\"lb-r\",    lbl:\"R\" },\r\n            \"Kotlin\":     { cls:\"lb-kt\",   lbl:\"Kotlin\" },\r\n            \"curl\":       { cls:\"lb-curl\", lbl:\"curl\" },\r\n            \"Lua\":        { cls:\"lb-lua\",  lbl:\"Lua\" }\r\n        },\r\n\r\n        modelsData: [\r\n            \/\/ --- TEXT \/ LLM  Local ---\r\n            { name:\"Llama-3.2-1B-Instruct\",          cat:\"Text \/ LLM\", type:\"Local \/ Browser\", tech:\"WebGPU (WebLLM \/ MLC)\",      size:\"~900 MB\",  use:\"Ultra-fast local chat & mobile-first assistants\", codeLangs:[\"Python\",\"JavaScript\",\"SQL\",\"Bash\"], sdkLangs:[\"JavaScript\"] },\r\n            { name:\"Llama-3.2-3B-Instruct\",          cat:\"Text \/ LLM\", type:\"Local \/ Browser\", tech:\"WebGPU (WebLLM)\",            size:\"~1.8 GB\",  use:\"Balanced quality chat & instruction following\", codeLangs:[\"Python\",\"JavaScript\",\"TypeScript\",\"Java\",\"SQL\"], sdkLangs:[\"JavaScript\"] },\r\n            { name:\"Llama-3.1-8B-Instruct (4-bit)\",  cat:\"Text \/ LLM\", type:\"Local \/ Browser\", tech:\"WebGPU (WebLLM)\",            size:\"~4.5 GB\",  use:\"High-quality reasoning, coding & long-context tasks\", codeLangs:[\"Python\",\"JavaScript\",\"TypeScript\",\"Java\",\"C++\",\"Go\",\"SQL\",\"Bash\"], sdkLangs:[\"JavaScript\"] },\r\n            { name:\"Phi-3.5-mini-3.8B (4-bit)\",      cat:\"Text \/ LLM\", type:\"Local \/ Browser\", tech:\"WebGPU + ONNX Runtime Web\",  size:\"~2.0 GB\",  use:\"Strong reasoning, code generation & multilingual\", codeLangs:[\"Python\",\"JavaScript\",\"TypeScript\",\"C#\",\"SQL\",\"Bash\"], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"Phi-4-mini (3.8B)\",              cat:\"Text \/ LLM\", type:\"Local \/ Browser\", tech:\"WebGPU \/ ONNX\",              size:\"~1.9 GB\",  use:\"Advanced local reasoning & coding (2026 upgrade)\", codeLangs:[\"Python\",\"JavaScript\",\"TypeScript\",\"C#\",\"C++\",\"SQL\"], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"Phi-2 (2.7B)\",                   cat:\"Text \/ LLM\", type:\"Local \/ Browser\", tech:\"ONNX Runtime Web\",           size:\"~1.4 GB\",  use:\"Small but capable code & reasoning tasks\", codeLangs:[\"Python\",\"JavaScript\",\"C#\",\"SQL\"], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"Gemma-2-2B-It\",                  cat:\"Text \/ LLM\", type:\"Local \/ Browser\", tech:\"WebGPU (WebLLM)\",            size:\"~1.5 GB\",  use:\"Balanced creative writing & fast responses\", codeLangs:[\"Python\",\"JavaScript\",\"Java\",\"SQL\"], sdkLangs:[\"JavaScript\"] },\r\n            { name:\"Gemma-2-9B-It (quantized)\",      cat:\"Text \/ LLM\", type:\"Local \/ Browser\", tech:\"WebGPU (WebLLM)\",            size:\"~4.8 GB\",  use:\"High-quality creative & technical writing\", codeLangs:[\"Python\",\"JavaScript\",\"TypeScript\",\"Java\",\"C++\",\"SQL\"], sdkLangs:[\"JavaScript\"] },\r\n            { name:\"Gemma-3-4B-It (quantized)\",      cat:\"Text \/ LLM\", type:\"Local \/ Browser\", tech:\"WebGPU (WebLLM)\",            size:\"~2.2 GB\",  use:\"2026 model: multimodal-ready, strong coding\", codeLangs:[\"Python\",\"JavaScript\",\"TypeScript\",\"Java\",\"C++\",\"Go\",\"SQL\"], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"SmolLM2-360M-Instruct\",          cat:\"Text \/ LLM\", type:\"Local \/ Browser\", tech:\"WASM + Transformers.js\",     size:\"~200 MB\",  use:\"Ultra-lightweight text generation for low-end devices\", codeLangs:[\"Python\",\"JavaScript\",\"SQL\"], sdkLangs:[\"JavaScript\"] },\r\n            { name:\"SmolLM2-1.7B-Instruct\",          cat:\"Text \/ LLM\", type:\"Local \/ Browser\", tech:\"Transformers.js \/ ONNX\",     size:\"~950 MB\",  use:\"Fast mid-tier chat, summaries & simple code\", codeLangs:[\"Python\",\"JavaScript\",\"TypeScript\",\"SQL\"], sdkLangs:[\"JavaScript\"] },\r\n            { name:\"Qwen2.5-0.5B-Instruct\",          cat:\"Text \/ LLM\", type:\"Local \/ Browser\", tech:\"WASM \/ ONNX\",                size:\"~350 MB\",  use:\"Efficient multilingual chat & translation\", codeLangs:[\"Python\",\"JavaScript\",\"SQL\"], sdkLangs:[\"JavaScript\"] },\r\n            { name:\"Qwen2.5-7B-Instruct (4-bit)\",    cat:\"Text \/ LLM\", type:\"Local \/ Browser\", tech:\"WebGPU \/ ONNX\",              size:\"~3.8 GB\",  use:\"Strong multilingual reasoning & coding\", codeLangs:[\"Python\",\"JavaScript\",\"TypeScript\",\"Java\",\"C++\",\"Go\",\"Rust\",\"SQL\"], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"Qwen2.5-Coder-7B (4-bit)\",       cat:\"Text \/ LLM\", type:\"Local \/ Browser\", tech:\"WebGPU (WebLLM)\",            size:\"~4.0 GB\",  use:\"Specialized local code generation & debugging\", codeLangs:[\"Python\",\"JavaScript\",\"TypeScript\",\"Java\",\"C++\",\"C#\",\"Go\",\"Rust\",\"PHP\",\"SQL\",\"Bash\"], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"Mistral-Nemo-12B (quantized)\",   cat:\"Text \/ LLM\", type:\"Local \/ Browser\", tech:\"WebGPU\",                     size:\"~6.2 GB\",  use:\"High-performance local assistant (2026)\", codeLangs:[\"Python\",\"JavaScript\",\"TypeScript\",\"Java\",\"C++\",\"Go\",\"Rust\",\"SQL\",\"Bash\"], sdkLangs:[\"JavaScript\"] },\r\n            { name:\"Mistral-7B-Instruct v0.3\",       cat:\"Text \/ LLM\", type:\"Local \/ Browser\", tech:\"WebGPU (WebLLM)\",            size:\"~4.1 GB\",  use:\"Reliable general instruction-following & coding\", codeLangs:[\"Python\",\"JavaScript\",\"TypeScript\",\"Java\",\"C++\",\"Go\",\"SQL\",\"Bash\"], sdkLangs:[\"JavaScript\"] },\r\n            { name:\"TinyLlama-1.1B-Chat-v1.0\",       cat:\"Text \/ LLM\", type:\"Local \/ Browser\", tech:\"WebGPU (WebLLM)\",            size:\"~650 MB\",  use:\"Super-fast basic chat on any device\", codeLangs:[\"Python\",\"JavaScript\",\"SQL\"], sdkLangs:[\"JavaScript\"] },\r\n            { name:\"DeepSeek-R1-1.5B (4-bit)\",       cat:\"Text \/ LLM\", type:\"Local \/ Browser\", tech:\"WebGPU \/ ONNX\",              size:\"~1.0 GB\",  use:\"Local chain-of-thought reasoning & math (2026)\", codeLangs:[\"Python\",\"JavaScript\",\"TypeScript\",\"C++\",\"SQL\"], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"DeepSeek-Coder-V2-Lite (4-bit)\", cat:\"Text \/ LLM\", type:\"Local \/ Browser\", tech:\"WebGPU (WebLLM)\",            size:\"~7.5 GB\",  use:\"Excellent local code completion & debugging\", codeLangs:[\"Python\",\"JavaScript\",\"TypeScript\",\"Java\",\"C++\",\"C#\",\"Go\",\"Rust\",\"PHP\",\"SQL\",\"Bash\"], sdkLangs:[\"JavaScript\"] },\r\n            { name:\"StarCoder2-3B\",                  cat:\"Text \/ LLM\", type:\"Local \/ Browser\", tech:\"Transformers.js \/ ONNX\",     size:\"~1.6 GB\",  use:\"Code-first: 80+ programming languages understood\", codeLangs:[\"Python\",\"JavaScript\",\"TypeScript\",\"Java\",\"C++\",\"C#\",\"Go\",\"Rust\",\"PHP\",\"Ruby\",\"SQL\",\"Bash\",\"Kotlin\"], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"CodeLlama-7B-Instruct (4-bit)\",  cat:\"Text \/ LLM\", type:\"Local \/ Browser\", tech:\"WebGPU (WebLLM)\",            size:\"~4.3 GB\",  use:\"Powerful local code generation & infilling\", codeLangs:[\"Python\",\"JavaScript\",\"TypeScript\",\"Java\",\"C++\",\"SQL\",\"Bash\"], sdkLangs:[\"JavaScript\"] },\r\n            { name:\"Hermes-3-Llama-3.1-8B (4-bit)\",  cat:\"Text \/ LLM\", type:\"Local \/ Browser\", tech:\"WebGPU (WebLLM)\",            size:\"~4.5 GB\",  use:\"Advanced function-calling & agentic tasks offline\", codeLangs:[\"Python\",\"JavaScript\",\"TypeScript\",\"Java\",\"Go\",\"Rust\",\"SQL\"], sdkLangs:[\"JavaScript\"] },\r\n            { name:\"Falcon-3-1B-Instruct\",           cat:\"Text \/ LLM\", type:\"Local \/ Browser\", tech:\"Transformers.js \/ ONNX\",     size:\"~600 MB\",  use:\"Efficient lightweight chat & reasoning\", codeLangs:[\"Python\",\"JavaScript\",\"SQL\"], sdkLangs:[\"JavaScript\"] },\r\n            { name:\"DistilBERT-base-uncased\",        cat:\"Text \/ LLM\", type:\"Local \/ Browser\", tech:\"WASM \/ Transformers.js\",     size:\"~250 MB\",  use:\"Fast text classification & sentiment analysis\", codeLangs:[], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"BERT-base-uncased\",              cat:\"Text \/ LLM\", type:\"Local \/ Browser\", tech:\"Transformers.js \/ ONNX\",     size:\"~420 MB\",  use:\"Token classification, QA & NLP fine-tuning\", codeLangs:[], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n\r\n            \/\/ --- TEXT \/ LLM  Cloud ---\r\n            { name:\"GPT-4o\",                         cat:\"Text \/ LLM\", type:\"Cloud API\", tech:\"OpenAI API\",            size:\"Cloud\", use:\"Advanced reasoning, coding, multi-modal chat\", codeLangs:[\"Python\",\"JavaScript\",\"TypeScript\",\"Java\",\"C++\",\"C#\",\"Go\",\"Rust\",\"PHP\",\"Ruby\",\"Swift\",\"SQL\",\"Bash\",\"R\",\"Kotlin\"], sdkLangs:[\"Python\",\"JavaScript\",\"C#\",\"Go\",\"Java\",\"curl\"] },\r\n            { name:\"GPT-4.1\",                        cat:\"Text \/ LLM\", type:\"Cloud API\", tech:\"OpenAI API\",            size:\"Cloud\", use:\"Flagship 2026 model: superior instruction following\", codeLangs:[\"Python\",\"JavaScript\",\"TypeScript\",\"Java\",\"C++\",\"C#\",\"Go\",\"Rust\",\"PHP\",\"Ruby\",\"Swift\",\"SQL\",\"Bash\",\"R\",\"Kotlin\"], sdkLangs:[\"Python\",\"JavaScript\",\"C#\",\"Go\",\"Java\",\"curl\"] },\r\n            { name:\"Claude 3.5 Sonnet\",              cat:\"Text \/ LLM\", type:\"Cloud API\", tech:\"Anthropic API\",         size:\"Cloud\", use:\"Fast, highly intelligent coding and writing\", codeLangs:[\"Python\",\"JavaScript\",\"TypeScript\",\"Java\",\"C++\",\"C#\",\"Go\",\"Rust\",\"Swift\",\"SQL\",\"Bash\",\"R\"], sdkLangs:[\"Python\",\"JavaScript\",\"TypeScript\",\"curl\"] },\r\n            { name:\"Claude 3.7 Opus\",                cat:\"Text \/ LLM\", type:\"Cloud API\", tech:\"Anthropic API\",         size:\"Cloud\", use:\"Highest-capability model for complex reasoning (2026)\", codeLangs:[\"Python\",\"JavaScript\",\"TypeScript\",\"Java\",\"C++\",\"C#\",\"Go\",\"Rust\",\"Swift\",\"PHP\",\"Ruby\",\"SQL\",\"Bash\",\"R\",\"Kotlin\"], sdkLangs:[\"Python\",\"JavaScript\",\"TypeScript\",\"curl\"] },\r\n            { name:\"Gemini 2.0 Flash\",               cat:\"Text \/ LLM\", type:\"Cloud API\", tech:\"Google Cloud \/ AI Studio\", size:\"Cloud\", use:\"Fastest Gemini for real-time & agentic workflows (2026)\", codeLangs:[\"Python\",\"JavaScript\",\"TypeScript\",\"Java\",\"C++\",\"Go\",\"Swift\",\"SQL\",\"Bash\",\"Kotlin\"], sdkLangs:[\"Python\",\"JavaScript\",\"Java\",\"Go\",\"curl\"] },\r\n            { name:\"Gemini 1.5 Pro\",                 cat:\"Text \/ LLM\", type:\"Cloud API\", tech:\"Google Cloud\",          size:\"Cloud\", use:\"Massive context window (up to 2M tokens)\", codeLangs:[\"Python\",\"JavaScript\",\"TypeScript\",\"Java\",\"C++\",\"Go\",\"Swift\",\"SQL\",\"Kotlin\"],  sdkLangs:[\"Python\",\"JavaScript\",\"Java\",\"Go\",\"curl\"] },\r\n            { name:\"Grok-3\",                         cat:\"Text \/ LLM\", type:\"Cloud API\", tech:\"xAI API\",               size:\"Cloud\", use:\"Real-time internet-aware reasoning & coding (2026)\", codeLangs:[\"Python\",\"JavaScript\",\"TypeScript\",\"Java\",\"C++\",\"Go\",\"Rust\",\"SQL\",\"Bash\"],     sdkLangs:[\"Python\",\"JavaScript\",\"curl\"] },\r\n            { name:\"DeepSeek-V3 (Cloud)\",            cat:\"Text \/ LLM\", type:\"Cloud API\", tech:\"DeepSeek API\",          size:\"Cloud\", use:\"Top open-weight model: code, math, reasoning\", codeLangs:[\"Python\",\"JavaScript\",\"TypeScript\",\"Java\",\"C++\",\"C#\",\"Go\",\"Rust\",\"SQL\"],        sdkLangs:[\"Python\",\"JavaScript\",\"curl\"] },\r\n            { name:\"Command R+ (Cohere)\",            cat:\"Text \/ LLM\", type:\"Cloud API\", tech:\"Cohere API\",            size:\"Cloud\", use:\"RAG-optimized enterprise search & summarization\", codeLangs:[\"Python\",\"JavaScript\",\"Java\",\"SQL\",\"Bash\"],                                     sdkLangs:[\"Python\",\"JavaScript\",\"Java\",\"Go\",\"curl\"] },\r\n            { name:\"Mistral Large 2\",                cat:\"Text \/ LLM\", type:\"Cloud API\", tech:\"Mistral API\",           size:\"Cloud\", use:\"Strong multilingual coding & function-calling\", codeLangs:[\"Python\",\"JavaScript\",\"TypeScript\",\"Java\",\"C++\",\"Go\",\"Rust\",\"SQL\",\"Bash\"],     sdkLangs:[\"Python\",\"JavaScript\",\"curl\"] },\r\n            { name:\"Llama 3.1-405B (Meta API)\",      cat:\"Text \/ LLM\", type:\"Cloud API\", tech:\"Meta \/ Together AI\",    size:\"Cloud\", use:\"Largest open-weight model for max capability\", codeLangs:[\"Python\",\"JavaScript\",\"TypeScript\",\"Java\",\"C++\",\"Go\",\"Rust\",\"SQL\",\"Bash\"],     sdkLangs:[\"Python\",\"JavaScript\",\"curl\"] },\r\n\r\n            \/\/ --- IMAGE  Local ---\r\n            { name:\"Stable Diffusion XL Turbo\",      cat:\"Image\", type:\"Local \/ Browser\", tech:\"WebGPU\",                size:\"~1.8 GB\", use:\"Real-time local 1-step image generation\", codeLangs:[], sdkLangs:[\"JavaScript\"] },\r\n            { name:\"Flux.1 Schnell (quantized)\",     cat:\"Image\", type:\"Local \/ Browser\", tech:\"ONNX Runtime Web\",      size:\"~4.2 GB\", use:\"High-quality local text-to-image (2026)\", codeLangs:[], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"Stable Diffusion 3 Medium (ONNX)\",cat:\"Image\", type:\"Local \/ Browser\", tech:\"ONNX Runtime Web\",     size:\"~5.1 GB\", use:\"Advanced local photorealistic image gen\", codeLangs:[], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n\r\n            \/\/ --- IMAGE  Cloud ---\r\n            { name:\"Midjourney v6.1\",                cat:\"Image\", type:\"Cloud API\", tech:\"Discord \/ REST\",              size:\"Cloud\", use:\"Ultra-realistic cinematic image generation\", codeLangs:[], sdkLangs:[\"Python\",\"JavaScript\",\"curl\"] },\r\n            { name:\"DALL-E 3\",                       cat:\"Image\", type:\"Cloud API\", tech:\"OpenAI API\",                  size:\"Cloud\", use:\"Prompt-accurate stylized image generation\", codeLangs:[], sdkLangs:[\"Python\",\"JavaScript\",\"C#\",\"Go\",\"curl\"] },\r\n            { name:\"Ideogram 2.0\",                   cat:\"Image\", type:\"Cloud API\", tech:\"Ideogram API\",                size:\"Cloud\", use:\"Images with perfect text rendering\", codeLangs:[], sdkLangs:[\"Python\",\"JavaScript\",\"curl\"] },\r\n            { name:\"Stable Diffusion 3.5 Large\",     cat:\"Image\", type:\"Cloud API\", tech:\"Stability AI API\",            size:\"Cloud\", use:\"Ultra high-res prompt-faithful images\", codeLangs:[], sdkLangs:[\"Python\",\"JavaScript\",\"curl\"] },\r\n            { name:\"Adobe Firefly 3\",                cat:\"Image\", type:\"Cloud API\", tech:\"Adobe API\",                   size:\"Cloud\", use:\"Commercially safe, brand-consistent generation\", codeLangs:[], sdkLangs:[\"Python\",\"JavaScript\",\"curl\"] },\r\n            { name:\"Recraft V3\",                     cat:\"Image\", type:\"Cloud API\", tech:\"Recraft API\",                 size:\"Cloud\", use:\"Vector\/SVG-native image & icon generation (2026)\", codeLangs:[], sdkLangs:[\"Python\",\"JavaScript\",\"curl\"] },\r\n\r\n            \/\/ --- VIDEO  Cloud ---\r\n            { name:\"Runway Gen-3 Alpha\",             cat:\"Video\", type:\"Cloud API\", tech:\"Runway API\",                  size:\"Cloud\", use:\"High-fidelity cinematic Text-to-Video\", codeLangs:[], sdkLangs:[\"Python\",\"JavaScript\",\"curl\"] },\r\n            { name:\"Kling AI 2.0\",                   cat:\"Video\", type:\"Cloud API\", tech:\"Kuaishou API\",                size:\"Cloud\", use:\"Ultra-realistic long-duration videos\", codeLangs:[], sdkLangs:[\"Python\",\"JavaScript\",\"curl\"] },\r\n            { name:\"Luma Dream Machine 1.6\",         cat:\"Video\", type:\"Cloud API\", tech:\"Luma API\",                    size:\"Cloud\", use:\"Fast high-quality Image-to-Video\", codeLangs:[], sdkLangs:[\"Python\",\"JavaScript\",\"curl\"] },\r\n            { name:\"Sora (OpenAI)\",                  cat:\"Video\", type:\"Cloud API\", tech:\"OpenAI API\",                  size:\"Cloud\", use:\"World-class long-form text-to-video generation\", codeLangs:[], sdkLangs:[\"Python\",\"JavaScript\",\"curl\"] },\r\n            { name:\"Pika 2.1\",                       cat:\"Video\", type:\"Cloud API\", tech:\"Pika API\",                    size:\"Cloud\", use:\"Fast & expressive video generation + editing\", codeLangs:[], sdkLangs:[\"Python\",\"JavaScript\",\"curl\"] },\r\n            { name:\"Veo 2 (Google)\",                 cat:\"Video\", type:\"Cloud API\", tech:\"Google Vertex AI\",            size:\"Cloud\", use:\"Cinematic HD video from text & image prompts (2026)\", codeLangs:[], sdkLangs:[\"Python\",\"JavaScript\",\"Java\",\"Go\",\"curl\"] },\r\n\r\n            \/\/ --- AUDIO \/ SPEECH  Local ---\r\n            { name:\"Whisper-Tiny.en\",                cat:\"Audio \/ Speech\", type:\"Local \/ Browser\", tech:\"WASM + WebGPU\",          size:\"~75 MB\",  use:\"Real-time English speech-to-text\", codeLangs:[], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"Whisper-Base (multilingual)\",    cat:\"Audio \/ Speech\", type:\"Local \/ Browser\", tech:\"Transformers.js \/ WASM\", size:\"~290 MB\", use:\"Fast multilingual speech recognition\", codeLangs:[], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"Whisper-Medium (quantized)\",     cat:\"Audio \/ Speech\", type:\"Local \/ Browser\", tech:\"ONNX \/ Transformers.js\", size:\"~1.2 GB\", use:\"Accurate multilingual speech-to-text\", codeLangs:[], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"Whisper-Large-v3 (quantized)\",   cat:\"Audio \/ Speech\", type:\"Local \/ Browser\", tech:\"ONNX Runtime Web\",       size:\"~3.1 GB\", use:\"Highest accuracy local STT with timestamps\", codeLangs:[], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"Piper TTS \/ VITS\",               cat:\"Audio \/ Speech\", type:\"Local \/ Browser\", tech:\"ONNX \/ WebAssembly\",      size:\"~30 MB\",  use:\"Fast offline text-to-speech\", codeLangs:[], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"Kokoro-82M TTS\",                 cat:\"Audio \/ Speech\", type:\"Local \/ Browser\", tech:\"Transformers.js\",         size:\"~165 MB\", use:\"Natural-sounding offline TTS (2026 favourite)\", codeLangs:[], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"Bark TTS (quantized)\",           cat:\"Audio \/ Speech\", type:\"Local \/ Browser\", tech:\"Transformers.js \/ ONNX\", size:\"~1.4 GB\", use:\"Realistic multi-speaker TTS with emotion\", codeLangs:[], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"MMS-TTS (Facebook)\",             cat:\"Audio \/ Speech\", type:\"Local \/ Browser\", tech:\"Transformers.js\",         size:\"~50 MB\",  use:\"1100+ language offline TTS (massively multilingual)\", codeLangs:[], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n\r\n            \/\/ --- AUDIO \/ SPEECH  Cloud ---\r\n            { name:\"ElevenLabs API\",                 cat:\"Audio \/ Speech\", type:\"Cloud API\", tech:\"ElevenLabs REST\",     size:\"Cloud\", use:\"Hyper-realistic TTS & voice cloning\", codeLangs:[], sdkLangs:[\"Python\",\"JavaScript\",\"curl\"] },\r\n            { name:\"Suno AI v4\",                     cat:\"Audio \/ Speech\", type:\"Cloud API\", tech:\"Suno Web \/ API\",      size:\"Cloud\", use:\"Full song generation with vocals & instruments\", codeLangs:[], sdkLangs:[\"Python\",\"JavaScript\",\"curl\"] },\r\n            { name:\"Udio\",                           cat:\"Audio \/ Speech\", type:\"Cloud API\", tech:\"Udio API\",            size:\"Cloud\", use:\"Studio-quality AI music generation (2026)\", codeLangs:[], sdkLangs:[\"Python\",\"JavaScript\",\"curl\"] },\r\n            { name:\"OpenAI Whisper API\",             cat:\"Audio \/ Speech\", type:\"Cloud API\", tech:\"OpenAI API\",          size:\"Cloud\", use:\"Managed Whisper with large-v3 quality at scale\", codeLangs:[], sdkLangs:[\"Python\",\"JavaScript\",\"C#\",\"Go\",\"curl\"] },\r\n\r\n            \/\/ --- VISION \/ UTILITIES  Local ---\r\n            { name:\"RMBG-2.0\",                       cat:\"Vision \/ Utils\", type:\"Local \/ Browser\", tech:\"ONNX Runtime Web\",      size:\"~90 MB\",  use:\"Real-time image background removal\", codeLangs:[], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"YOLOv10 (Web Export)\",           cat:\"Vision \/ Utils\", type:\"Local \/ Browser\", tech:\"ONNX Runtime Web\",      size:\"~15 MB\",  use:\"Real-time object detection & tracking\", codeLangs:[], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"YOLOv11 (ONNX Web)\",             cat:\"Vision \/ Utils\", type:\"Local \/ Browser\", tech:\"ONNX Runtime Web\",      size:\"~20 MB\",  use:\"State-of-art real-time detection (2026)\", codeLangs:[], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"MediaPipe Face Landmarker\",      cat:\"Vision \/ Utils\", type:\"Local \/ Browser\", tech:\"WebGPU\",                size:\"~8 MB\",   use:\"Real-time face detection & 478-point landmarks\", codeLangs:[], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"MediaPipe Hand Landmarker\",      cat:\"Vision \/ Utils\", type:\"Local \/ Browser\", tech:\"WebGPU\",                size:\"~12 MB\",  use:\"Real-time hand tracking & gesture recognition\", codeLangs:[], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"MediaPipe Pose Landmarker\",      cat:\"Vision \/ Utils\", type:\"Local \/ Browser\", tech:\"WebGPU\",                size:\"~14 MB\",  use:\"Full body pose estimation & skeleton tracking\", codeLangs:[], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"Tesseract.js\",                   cat:\"Vision \/ Utils\", type:\"Local \/ Browser\", tech:\"WebAssembly (WASM)\",   size:\"~25 MB\",  use:\"Offline OCR \u2014 text extraction from images\", codeLangs:[], sdkLangs:[\"JavaScript\"] },\r\n            { name:\"MobileNetV3-Small\",              cat:\"Vision \/ Utils\", type:\"Local \/ Browser\", tech:\"TensorFlow.js\",         size:\"~15 MB\",  use:\"Lightweight image classification (1000 classes)\", codeLangs:[], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"Swin-Tiny Transformer\",          cat:\"Vision \/ Utils\", type:\"Local \/ Browser\", tech:\"Transformers.js\",       size:\"~45 MB\",  use:\"High-accuracy image classification\", codeLangs:[], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"MiDaS Depth Estimator\",          cat:\"Vision \/ Utils\", type:\"Local \/ Browser\", tech:\"ONNX \/ Transformers.js\",size:\"~25 MB\",  use:\"Depth estimation from 2D images\", codeLangs:[], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"DepthAnything v2 (small)\",       cat:\"Vision \/ Utils\", type:\"Local \/ Browser\", tech:\"Transformers.js\",       size:\"~98 MB\",  use:\"Best-in-class monocular depth estimation (2026)\", codeLangs:[], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"SAM2 Tiny (Segment Anything)\",   cat:\"Vision \/ Utils\", type:\"Local \/ Browser\", tech:\"ONNX Runtime Web\",      size:\"~50 MB\",  use:\"Real-time interactive image segmentation (2026)\", codeLangs:[], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"CLIP ViT-B\/32\",                  cat:\"Vision \/ Utils\", type:\"Local \/ Browser\", tech:\"Transformers.js\",       size:\"~150 MB\", use:\"Zero-shot image classification & image-text search\", codeLangs:[], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"Florence-2 Base (local)\",        cat:\"Vision \/ Utils\", type:\"Local \/ Browser\", tech:\"Transformers.js \/ ONNX\",size:\"~450 MB\", use:\"Multi-task vision: captioning, detection, OCR\", codeLangs:[], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"NSFWJS \/ NudeNet\",               cat:\"Vision \/ Utils\", type:\"Local \/ Browser\", tech:\"TensorFlow.js\",         size:\"~5 MB\",   use:\"Offline image moderation & content filtering\", codeLangs:[], sdkLangs:[\"JavaScript\"] },\r\n\r\n            \/\/ --- EMBEDDING \/ RAG  Local ---\r\n            { name:\"BGE-small-en-v1.5\",              cat:\"Embedding\", type:\"Local \/ Browser\", tech:\"Transformers.js (WASM)\", size:\"~120 MB\", use:\"Semantic embeddings & RAG vector search\", codeLangs:[], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"BGE-large-en-v1.5\",              cat:\"Embedding\", type:\"Local \/ Browser\", tech:\"Transformers.js (WASM)\", size:\"~1.3 GB\", use:\"Highest quality BAAI embeddings for RAG\", codeLangs:[], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"all-MiniLM-L6-v2\",               cat:\"Embedding\", type:\"Local \/ Browser\", tech:\"Transformers.js (WASM)\", size:\"~80 MB\",  use:\"Lightning-fast sentence embeddings for similarity\", codeLangs:[], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"all-MiniLM-L12-v2\",              cat:\"Embedding\", type:\"Local \/ Browser\", tech:\"Transformers.js (WASM)\", size:\"~130 MB\", use:\"Higher accuracy semantic sentence embeddings\", codeLangs:[], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"nomic-embed-text-v1.5\",          cat:\"Embedding\", type:\"Local \/ Browser\", tech:\"Transformers.js\",        size:\"~140 MB\", use:\"High-quality long-context embeddings (2026 top pick)\", codeLangs:[], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"LaBSE\",                          cat:\"Embedding\", type:\"Local \/ Browser\", tech:\"Transformers.js\",        size:\"~480 MB\", use:\"Multilingual sentence embeddings (109 languages)\", codeLangs:[], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"E5-large-v2\",                    cat:\"Embedding\", type:\"Local \/ Browser\", tech:\"Transformers.js (WASM)\", size:\"~1.3 GB\", use:\"Instruction-following embeddings for RAG\", codeLangs:[], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n            { name:\"Jina Embeddings v3 (local)\",     cat:\"Embedding\", type:\"Local \/ Browser\", tech:\"Transformers.js\",        size:\"~570 MB\", use:\"Task-specific embeddings with LoRA adapters (2026)\", codeLangs:[], sdkLangs:[\"JavaScript\",\"Python\"] },\r\n\r\n            \/\/ --- EMBEDDING  Cloud ---\r\n            { name:\"OpenAI text-embedding-3-large\",  cat:\"Embedding\", type:\"Cloud API\", tech:\"OpenAI API\",                  size:\"Cloud\", use:\"Best OpenAI embedding model (3072-dim)\", codeLangs:[], sdkLangs:[\"Python\",\"JavaScript\",\"C#\",\"Go\",\"Java\",\"curl\"] },\r\n            { name:\"Cohere Embed v3\",                cat:\"Embedding\", type:\"Cloud API\", tech:\"Cohere API\",                  size:\"Cloud\", use:\"Multilingual embeddings optimized for search & RAG\", codeLangs:[], sdkLangs:[\"Python\",\"JavaScript\",\"Java\",\"Go\",\"curl\"] }\r\n        ],\r\n\r\n        init: function() {\r\n            this.cacheDOM();\r\n            this.attachEvents();\r\n            this.renderTable(this.modelsData);\r\n        },\r\n\r\n        cacheDOM: function() {\r\n            this.UI = {\r\n                layout:      document.getElementById('pth-ai-models-dir-layout'),\r\n                sidebar:     document.getElementById('pth-ai-models-dir-sidebar'),\r\n                fwBtn:       document.getElementById('pth-fw-btn-ai-dir'),\r\n                sampleBtn:   document.getElementById('pth-load-sample-btn'),\r\n                searchInput: document.getElementById('pth-dir-search'),\r\n                filterCat:   document.getElementById('pth-filter-cat'),\r\n                filterType:  document.getElementById('pth-filter-type'),\r\n                filterLang:  document.getElementById('pth-filter-lang'),\r\n                resetBtn:    document.getElementById('pth-reset-filters-btn'),\r\n                tableBody:   document.getElementById('pth-table-body'),\r\n                emptyState:  document.getElementById('pth-empty-state'),\r\n                resultCount: document.getElementById('pth-result-count'),\r\n                sortHeaders: document.querySelectorAll('.pth-models-table th.sortable'),\r\n                exportCsvBtn:document.getElementById('pth-export-csv-btn'),\r\n                printBtn:    document.getElementById('pth-print-btn'),\r\n                copyJsonBtn: document.getElementById('pth-copy-json-btn'),\r\n                quickFilterBtns: document.querySelectorAll('.pth-quick-filter-btn'),\r\n                toast:       document.getElementById('pth-dir-toast')\r\n            };\r\n        },\r\n\r\n        showToast: function(msg) {\r\n            if (!this.UI.toast) return;\r\n            this.UI.toast.innerHTML = msg;\r\n            this.UI.toast.classList.add('show');\r\n            setTimeout(() => { this.UI.toast.classList.remove('show'); }, 3000);\r\n        },\r\n\r\n        toggleWidth: function() {\r\n            if (this.UI.layout.classList.contains('pth-full-width')) {\r\n                this.UI.layout.classList.remove('pth-full-width');\r\n                this.UI.sidebar.style.display = 'block';\r\n                this.UI.fwBtn.innerHTML = '\u26f6 Full Width';\r\n            } else {\r\n                this.UI.layout.classList.add('pth-full-width');\r\n                this.UI.sidebar.style.display = 'none';\r\n                this.UI.fwBtn.innerHTML = '\u25e8 Split View';\r\n            }\r\n        },\r\n\r\n        attachEvents: function() {\r\n            const self = this;\r\n            if (this.UI.fwBtn)    this.UI.fwBtn.addEventListener('click', () => self.toggleWidth());\r\n            if (this.UI.sampleBtn)this.UI.sampleBtn.addEventListener('click', () => self.loadSample());\r\n            \r\n            this.UI.searchInput.addEventListener('keyup', () => self.filterTable());\r\n            this.UI.filterCat.addEventListener('change', () => self.filterTable());\r\n            this.UI.filterType.addEventListener('change', () => self.filterTable());\r\n            this.UI.filterLang.addEventListener('change', () => self.filterTable());\r\n            this.UI.resetBtn.addEventListener('click', () => self.resetFilters());\r\n            \r\n            this.UI.sortHeaders.forEach(th => {\r\n                th.addEventListener('click', function() { self.sortTable(this.dataset.sort); });\r\n            });\r\n            \r\n            this.UI.exportCsvBtn.addEventListener('click', () => self.exportCSV());\r\n            this.UI.printBtn.addEventListener('click', () => { window.print(); });\r\n            this.UI.copyJsonBtn.addEventListener('click', () => self.copyJSON());\r\n\r\n            this.UI.quickFilterBtns.forEach(btn => {\r\n                btn.addEventListener('click', function() {\r\n                    self.quickFilter(this.dataset.cat, this.dataset.type, this.dataset.lang);\r\n                });\r\n            });\r\n        },\r\n\r\n        makeBadges: function(arr) {\r\n            return arr.map(l => {\r\n                const d = this.LB[l] || { cls:\"lb-na\", lbl: l };\r\n                return '\\x3Cspan class=\"pth-lb ' + d.cls + '\"\\x3E' + d.lbl + '\\x3C\/span\\x3E';\r\n            }).join('');\r\n        },\r\n\r\n        buildLangCell: function(row) {\r\n            let html = '';\r\n            if (row.codeLangs && row.codeLangs.length > 0) {\r\n                html += '\\x3Cdiv class=\"pth-langs-section\"\\x3E\\x3Cspan class=\"pth-langs-label\"\\x3E\\uD83D\\uDDA5\\uFE0F Handles code in\\x3C\/span\\x3E\\x3Cdiv class=\"pth-lang-wrap\"\\x3E' + this.makeBadges(row.codeLangs) + '\\x3C\/div\\x3E\\x3C\/div\\x3E';\r\n            } else if (row.cat === 'Text \/ LLM') {\r\n                html += '\\x3Cdiv class=\"pth-langs-section\"\\x3E\\x3Cspan class=\"pth-langs-label\"\\x3E\\uD83D\\uDDA5\\uFE0F Handles code in\\x3C\/span\\x3E\\x3Cdiv class=\"pth-lang-wrap\"\\x3E\\x3Cspan class=\"pth-lb lb-na\"\\x3EClassification only\\x3C\/span\\x3E\\x3C\/div\\x3E\\x3C\/div\\x3E';\r\n            } else {\r\n                html += '\\x3Cdiv class=\"pth-langs-section\"\\x3E\\x3Cspan class=\"pth-langs-label\"\\x3E\\uD83D\\uDDA5\\uFE0F Handles code in\\x3C\/span\\x3E\\x3Cdiv class=\"pth-lang-wrap\"\\x3E\\x3Cspan class=\"pth-lb lb-na\"\\x3EN\/A\\x3C\/span\\x3E\\x3C\/div\\x3E\\x3C\/div\\x3E';\r\n            }\r\n            if (row.sdkLangs && row.sdkLangs.length > 0) {\r\n                html += '\\x3Chr class=\"pth-lang-divider\"\\x3E\\x3Cdiv class=\"pth-langs-section\"\\x3E\\x3Cspan class=\"pth-langs-label\"\\x3E\\uD83D\\uDD17 Connect \/ SDK\\x3C\/span\\x3E\\x3Cdiv class=\"pth-lang-wrap\"\\x3E' + this.makeBadges(row.sdkLangs) + '\\x3C\/div\\x3E\\x3C\/div\\x3E';\r\n            }\r\n            return html;\r\n        },\r\n\r\n        renderTable: function(data) {\r\n            this.UI.tableBody.innerHTML = '';\r\n            if (this.UI.resultCount) this.UI.resultCount.textContent = data.length;\r\n            if (data.length === 0) {\r\n                this.UI.emptyState.style.display = 'block';\r\n                return;\r\n            }\r\n            this.UI.emptyState.style.display = 'none';\r\n\r\n            data.forEach(row => {\r\n                const tr = document.createElement('tr');\r\n                const typeBadge = row.type === 'Local \/ Browser'\r\n                    ? '\\x3Cspan class=\"pth-tag pth-tag-local\"\\x3E\\uD83D\\uDFE2 Offline Browser Capable\\x3C\/span\\x3E'\r\n                    : '\\x3Cspan class=\"pth-tag pth-tag-cloud\"\\x3E\\uD83D\\uDD34 Cloud Only (Internet Req.)\\x3C\/span\\x3E';\r\n                const catBadge = '\\x3Cspan class=\"pth-tag pth-tag-cat\"\\x3E' + row.cat + '\\x3C\/span\\x3E';\r\n                \r\n                tr.innerHTML = \r\n                    '\\x3Ctd\\x3E\\x3Cstrong style=\"font-size:0.87rem;\"\\x3E' + row.name + '\\x3C\/strong\\x3E\\x3C\/td\\x3E' +\r\n                    '\\x3Ctd\\x3E' + catBadge + '\\x3C\/td\\x3E' +\r\n                    '\\x3Ctd\\x3E\\x3Cdiv style=\"margin-bottom:7px;\"\\x3E' + typeBadge + '\\x3C\/div\\x3E\\x3Cdiv style=\"font-size:0.78rem; color:#64748b;\"\\x3E\\x3Cstrong\\x3ETech:\\x3C\/strong\\x3E ' + row.tech + '\\x3C\/div\\x3E\\x3Cdiv style=\"font-size:0.78rem; color:#64748b;\"\\x3E\\x3Cstrong\\x3ESize:\\x3C\/strong\\x3E ' + row.size + '\\x3C\/div\\x3E\\x3C\/td\\x3E' +\r\n                    '\\x3Ctd\\x3E' + this.buildLangCell(row) + '\\x3C\/td\\x3E' +\r\n                    '\\x3Ctd style=\"line-height:1.5; font-size:0.85rem;\"\\x3E' + row.use + '\\x3C\/td\\x3E';\r\n                \r\n                this.UI.tableBody.appendChild(tr);\r\n            });\r\n        },\r\n\r\n        filterTable: function() {\r\n            const searchTerm = this.UI.searchInput.value.toLowerCase().trim();\r\n            const catFilter  = this.UI.filterCat.value;\r\n            const typeFilter = this.UI.filterType.value;\r\n            const langFilter = this.UI.filterLang.value;\r\n            \r\n            const filtered = this.modelsData.filter(m => {\r\n                const allLangs = [...(m.codeLangs||[]), ...(m.sdkLangs||[])];\r\n                const matchesSearch = !searchTerm ||\r\n                    m.name.toLowerCase().includes(searchTerm) ||\r\n                    m.tech.toLowerCase().includes(searchTerm) ||\r\n                    m.use.toLowerCase().includes(searchTerm) ||\r\n                    allLangs.some(l => l.toLowerCase().includes(searchTerm));\r\n                const matchesCat  = !catFilter  || m.cat  === catFilter;\r\n                const matchesType = !typeFilter || m.type === typeFilter;\r\n                const matchesLang = !langFilter || allLangs.some(l => l === langFilter);\r\n                \r\n                return matchesSearch && matchesCat && matchesType && matchesLang;\r\n            });\r\n            this.renderTable(filtered);\r\n        },\r\n\r\n        sortTable: function(colKey) {\r\n            if (this.currentSort.col === colKey) {\r\n                this.currentSort.dir *= -1;\r\n            } else {\r\n                this.currentSort.col = colKey;\r\n                this.currentSort.dir = 1;\r\n            }\r\n            \r\n            this.modelsData.sort((a, b) => {\r\n                let vA = a[colKey] || '';\r\n                let vB = b[colKey] || '';\r\n                return this.currentSort.dir * vA.toString().localeCompare(vB.toString());\r\n            });\r\n            \r\n            this.UI.sortHeaders.forEach(th => {\r\n                let base = th.innerText.replace(\/ [\u2b06\u2b07\u2195]$\/, '');\r\n                th.innerText = base + (th.dataset.sort === colKey ? (this.currentSort.dir === 1 ? ' \u2b06' : ' \u2b07') : ' \u2195');\r\n            });\r\n            \r\n            this.filterTable();\r\n        },\r\n\r\n        quickFilter: function(cat, type, lang) {\r\n            this.UI.filterCat.value  = cat  || '';\r\n            this.UI.filterType.value = type || '';\r\n            this.UI.filterLang.value = lang || '';\r\n            this.UI.searchInput.value = '';\r\n            this.filterTable();\r\n            this.showToast(\"\ud83d\udd0d Filter applied\");\r\n        },\r\n\r\n        resetFilters: function() {\r\n            this.UI.searchInput.value = '';\r\n            this.UI.filterCat.value   = '';\r\n            this.UI.filterType.value  = '';\r\n            this.UI.filterLang.value  = '';\r\n            this.filterTable();\r\n            this.showToast(\"\ud83d\udd04 All filters reset\");\r\n        },\r\n\r\n        loadSample: function() {\r\n            this.UI.filterType.value = 'Local \/ Browser';\r\n            this.UI.filterCat.value  = 'Text \/ LLM';\r\n            this.UI.filterLang.value = 'Python';\r\n            this.UI.searchInput.value = '';\r\n            this.filterTable();\r\n            this.showToast(\"\ud83d\udca1 Showing: Python-connectable Local LLMs\");\r\n        },\r\n\r\n        exportCSV: function() {\r\n            let csv = \"Model Name,Category,Deployment,Technology,Size,Handles Code In,Connect SDK,Primary Use Case\\n\";\r\n            this.modelsData.forEach(m => {\r\n                const code = (m.codeLangs||[]).join(' | ');\r\n                const sdk  = (m.sdkLangs||[]).join(' | ');\r\n                csv += `\"${m.name}\",\"${m.cat}\",\"${m.type}\",\"${m.tech}\",\"${m.size}\",\"${code}\",\"${sdk}\",\"${m.use}\"\\n`;\r\n            });\r\n            const blob = new Blob([csv], { type: 'text\/csv' });\r\n            const url  = URL.createObjectURL(blob);\r\n            const a    = document.createElement('a');\r\n            a.href = url;\r\n            a.download = 'PrimeToolHub-AI-Models-Directory-2026.csv';\r\n            a.click();\r\n            URL.revokeObjectURL(url);\r\n            this.showToast('\ud83d\udce5 CSV Exported with Language columns!');\r\n        },\r\n\r\n        copyJSON: function() {\r\n            const json = JSON.stringify(this.modelsData, null, 2);\r\n            navigator.clipboard.writeText(json).then(() => {\r\n                this.showToast('\ud83d\udccb Full JSON copied (includes language data)!');\r\n            });\r\n        }\r\n    };\r\n\r\n    \/\/ Bulletproof Initializer\r\n    function runInit() {\r\n        if (!window.pthAiModelsDirApp.initialized && document.getElementById('pth-ai-models-dir-wrapper')) {\r\n            window.pthAiModelsDirApp.initialized = true;\r\n            window.pthAiModelsDirApp.init();\r\n        }\r\n    }\r\n\r\n    if (document.readyState === 'complete' || document.readyState === 'interactive') {\r\n        setTimeout(runInit, 1);\r\n    } else {\r\n        document.addEventListener('DOMContentLoaded', runInit);\r\n        window.addEventListener('load', runInit);\r\n    }\r\n})();\r\n<\/script>\n\n\n\n<style>\n.pth-bai-intro{max-width:1480px;margin:30px auto 10px;padding:26px 28px;background:#f8fafc;border:1px solid #e2e8f0;border-left:4px solid #2563eb;border-radius:14px;box-sizing:border-box}\n.pth-bai-intro h2{font-size:1.6rem;font-weight:800;color:#0f172a;margin:0 0 12px}\n.pth-bai-intro p{margin:0;color:#1e293b;font-weight:500;line-height:1.7}\n<\/style>\n<div class=\"pth-bai-intro\">\n  <h2>Picking a model you can actually run, not just read about<\/h2>\n  <p>There are hundreds of open models now, but only some of them fit inside a browser tab and run on the hardware you actually have. This directory exists to answer one narrow question quickly: given your device, your language stack, and whether you need it offline, which browser AI models are worth trying first? The sections below show how to read the list, walk through a few real searches, and are honest about what this reference will and will not do for you.<\/p>\n<\/div>\n\n\n\n<p class=\"wp-block-paragraph\">Last updated: July 2026<\/p>\n\n\n<!-- schoolictTool-il:2717-6677 -->\n\n<p class=\"schoolictTool-il wp-block-paragraph\">Learn how it works: <a href=\"https:\/\/schoolict.net\/tools\/how-browser-ai-code-tools-work\/\">Browser AI Coding Tools and BYOK Privacy Basics<\/a>.<\/p>\n\n<!-- \/schoolictTool-il:2717-6677 -->\n\n\n<h2 id=\"\ud83d\udd34-how-to-read-the-directory-without-getting-lost\" class=\"wp-block-heading\">\ud83d\udd34&nbsp;<strong>How to read the directory without getting lost<\/strong><\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Every row is one model, and the columns are the things that actually decide whether it will work for you. The category tells you what it does (text chat, speech-to-text, image, embeddings). The tech column is the engine it needs, usually WebGPU for GPU-accelerated work or WebAssembly for a CPU-only fallback. The size is the rough download you should expect the first time you load it. And the language columns tell you which programming languages you can drive it from, split into the code languages the model touches and the SDK languages you would actually write in the browser.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The filters at the top let you narrow all of that at once. Search by name, restrict to a category, filter by offline status, or pick a technology. The practical trick is to filter by the one constraint you cannot change first, then sort by size. Your hardware and your language stack are fixed; the model is the variable you are shopping for.<\/p>\n\n\n\n<h2 id=\"\ud83d\udfe1-a-few-real-searches-and-what-comes-back\" class=\"wp-block-heading\">\ud83d\udfe1&nbsp;<strong>A few real searches, and what comes back<\/strong><\/h2>\n\n\n<figure class=\"pth-article-figure pth-img-left\" style=\"float:left; width:700px; max-width:100%; margin:4px 28px 16px 0; clear:left;\"><img decoding=\"async\" src=\"https:\/\/schoolict.net\/tools\/wp-content\/uploads\/2026\/07\/browser-ai-filter-workflow-800x447.jpeg\" alt=\"Filtering a browser AI models directory by category, size and language\" width=\"700\" height=\"394\" loading=\"lazy\" data-no-lazy=\"1\" class=\"pth-article-img\" style=\"width:100%;height:auto;display:block;border-radius:10px;border:1px solid #e2e8f0;\"><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">Say you want a chat assistant that runs on a mid-range phone. Filter the category to text or LLM, then sort by size ascending. The top of that list will be something like a one-billion-parameter model around 900&nbsp;MB on WebGPU, tagged for fast local chat. Input: category equals text, sort by smallest. Output: a short list of models light enough that a phone can load them, each with the exact SDK language you would use to wire it up.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">A second common case: you need speech-to-text but the audio is sensitive and cannot be uploaded. Filter the category to audio and look for a Whisper-family entry with an offline tag. What you get back is a handful of transcription models with their sizes, so you can trade accuracy against download weight before you commit. A third case is a hard language constraint. If your project is pure JavaScript with no Python backend, filter by SDK language for JavaScript, and the list collapses to only the models you can integrate without standing up a server.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Once you have a candidate, the next step is a tool that actually runs it. If you are prototyping, the <a href=\"https:\/\/schoolict.net\/tools\/ai-code-studio\/\" rel=\"noreferrer noopener\" target=\"_blank\">AI Code Studio<\/a> lets you test model calls in the browser, and for the speech case the <a href=\"https:\/\/schoolict.net\/tools\/ai-offline-transcription\/\" rel=\"noreferrer noopener\" target=\"_blank\">offline transcription tool<\/a> already wires a Whisper model up for you. This directory is the shortlist; those tools are where you put a model to work.<\/p>\n\n\n\n<h2 id=\"\ud83d\udfe2-what-this-directory-will-not-do-for-you\" class=\"wp-block-heading\">\ud83d\udfe2&nbsp;<strong>What this directory will not do for you<\/strong><\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Being clear about the edges saves you time. This is a reference and comparison list, not a runtime. It does not run the models for you, and clicking a row will not start an inference. The sizes are approximate; the real download depends on the exact build and quantization you end up using, and can be larger or smaller than the figure shown. WebGPU support also varies by browser, so a model tagged WebGPU will want a recent Chrome or Edge, and older browsers fall back to the slower WebAssembly path or will not run it at all.<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>\ud83d\udd35 <strong>Hardware still rules.<\/strong> A large multi-billion-parameter model listed here can be downloaded, but a phone or a laptop without a capable GPU will struggle to run it at a usable speed.<\/li>\n\n\n\n<li>\ud83d\udfe0 <strong>The list is curated, not exhaustive.<\/strong> It covers the models worth trying in a browser today, not every model ever published. New ones appear faster than any single list can track.<\/li>\n\n\n\n<li>\ud83d\udfe3 <strong>First load is the cost.<\/strong> The size column is a one-time download that the browser then caches, not something you pay on every visit. Plan your first run on a good connection.<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">If you want the reasoning behind all of this, why the same model ships in three different sizes, how WebGPU and WebAssembly split the work, and why running locally keeps your data private, that is covered in the companion guide on <a href=\"https:\/\/schoolict.net\/tools\/how-browser-ai-models-work\/\" rel=\"noreferrer noopener\" target=\"_blank\">how browser AI models actually work<\/a> <em>[confirm slug \u2014 new theory post]<\/em>.<\/p>\n\n\n\n<style>\n.pth-faq-section{margin:50px auto 40px;font-family:inherit;max-width:1480px;padding:0 20px;box-sizing:border-box}\n.pth-faq-header{font-size:1.8rem;font-weight:800;color:#0f172a;margin-bottom:25px;border-bottom:2px solid #e2e8f0;padding-bottom:10px;display:flex;align-items:center;gap:10px}\n.pth-faq-grid{display:grid;grid-template-columns:1fr;gap:20px}\n@media(min-width:768px){.pth-faq-grid{grid-template-columns:repeat(2,1fr)}}\n@media(min-width:1024px){.pth-faq-grid{grid-template-columns:repeat(3,1fr)}}\n.pth-faq-card{background:#f8fafc;padding:24px;border-radius:12px;border:1px solid #e2e8f0;transition:transform .2s ease;break-inside:avoid}\n.pth-faq-card:hover{transform:translateY(-3px);box-shadow:0 4px 12px rgba(0,0,0,.05)}\n.pth-faq-q{color:#0f172a;font-size:1rem;font-weight:700;margin:0 0 12px;line-height:1.4}\n.pth-faq-a{margin:0;font-size:.95rem;color:#1e293b;line-height:1.6;font-weight:500}\n<\/style>\n<div class=\"pth-faq-section\">\n  <div class=\"pth-faq-header\">\u2753 Frequently Asked Questions<\/div>\n  <div class=\"pth-faq-grid\">\n    <div class=\"pth-faq-card\"><p class=\"pth-faq-q\">Does this directory run the models?<\/p><p class=\"pth-faq-a\">No. It is a reference and comparison list. It shows each model&#8217;s engine, size, and integration languages so you can shortlist, then you run your pick in a separate tool or your own code.<\/p><\/div>\n    <div class=\"pth-faq-card\"><p class=\"pth-faq-q\">Are the model sizes exact?<\/p><p class=\"pth-faq-a\">They are approximate. The real download depends on the exact build and quantization you use, so treat the figure as a guide for planning, not a precise number.<\/p><\/div>\n    <div class=\"pth-faq-card\"><p class=\"pth-faq-q\">What is the difference between WebGPU and WebAssembly here?<\/p><p class=\"pth-faq-a\">WebGPU runs the model on your graphics hardware and is much faster; WebAssembly runs it on the CPU as a fallback. A WebGPU model wants a recent Chrome or Edge to reach full speed.<\/p><\/div>\n    <div class=\"pth-faq-card\"><p class=\"pth-faq-q\">Can my phone run these?<\/p><p class=\"pth-faq-a\">The small models, yes. Filter the category you want and sort by size; the lightest entries are designed to load on mobile. Large multi-billion-parameter models are not realistic on a phone.<\/p><\/div>\n    <div class=\"pth-faq-card\"><p class=\"pth-faq-q\">Do I need to write code to use a model?<\/p><p class=\"pth-faq-a\">Yes, or use a tool that already wraps one. The SDK language column tells you what you would write, and most browser models integrate through a JavaScript SDK.<\/p><\/div>\n    <div class=\"pth-faq-card\"><p class=\"pth-faq-q\">Is my data private if I run a listed model?<\/p><p class=\"pth-faq-a\">When a model runs in your browser, the input stays on your device and is not sent to a server. The companion theory guide explains exactly why that holds.<\/p><\/div>\n    <div class=\"pth-faq-card\"><p class=\"pth-faq-q\">How current is the list?<\/p><p class=\"pth-faq-a\">It is curated and refreshed periodically to cover models worth trying in a browser now. It is intentionally selective rather than a complete index of everything published.<\/p><\/div>\n    <div class=\"pth-faq-card\"><p class=\"pth-faq-q\">Which model should a beginner start with?<\/p><p class=\"pth-faq-a\">Pick a small WebGPU text model in the one-billion-parameter range. It loads quickly, runs on modest hardware, and is enough to learn how in-browser inference behaves before you scale up.<\/p><\/div>\n  <\/div>\n<\/div>\n\n\n\n<script type=\"application\/ld+json\">\n{\n  \"@context\": \"https:\/\/schema.org\",\n  \"@type\": \"FAQPage\",\n  \"mainEntity\": [\n    {\"@type\":\"Question\",\"name\":\"Does this directory run the models?\",\"acceptedAnswer\":{\"@type\":\"Answer\",\"text\":\"No. It is a reference and comparison list. It shows each model's engine, size, and integration languages so you can shortlist, then you run your pick in a separate tool or your own code.\"}},\n    {\"@type\":\"Question\",\"name\":\"Are the model sizes exact?\",\"acceptedAnswer\":{\"@type\":\"Answer\",\"text\":\"They are approximate. The real download depends on the exact build and quantization you use, so treat the figure as a guide for planning, not a precise number.\"}},\n    {\"@type\":\"Question\",\"name\":\"What is the difference between WebGPU and WebAssembly here?\",\"acceptedAnswer\":{\"@type\":\"Answer\",\"text\":\"WebGPU runs the model on your graphics hardware and is much faster; WebAssembly runs it on the CPU as a fallback. A WebGPU model wants a recent Chrome or Edge to reach full speed.\"}},\n    {\"@type\":\"Question\",\"name\":\"Can my phone run these?\",\"acceptedAnswer\":{\"@type\":\"Answer\",\"text\":\"The small models, yes. Filter the category you want and sort by size; the lightest entries are designed to load on mobile. Large multi-billion-parameter models are not realistic on a phone.\"}},\n    {\"@type\":\"Question\",\"name\":\"Do I need to write code to use a model?\",\"acceptedAnswer\":{\"@type\":\"Answer\",\"text\":\"Yes, or use a tool that already wraps one. The SDK language column tells you what you would write, and most browser models integrate through a JavaScript SDK.\"}},\n    {\"@type\":\"Question\",\"name\":\"Is my data private if I run a listed model?\",\"acceptedAnswer\":{\"@type\":\"Answer\",\"text\":\"When a model runs in your browser, the input stays on your device and is not sent to a server. The companion theory guide explains exactly why that holds.\"}},\n    {\"@type\":\"Question\",\"name\":\"How current is the list?\",\"acceptedAnswer\":{\"@type\":\"Answer\",\"text\":\"It is curated and refreshed periodically to cover models worth trying in a browser now. It is intentionally selective rather than a complete index of everything published.\"}},\n    {\"@type\":\"Question\",\"name\":\"Which model should a beginner start with?\",\"acceptedAnswer\":{\"@type\":\"Answer\",\"text\":\"Pick a small WebGPU text model in the one-billion-parameter range. It loads quickly, runs on modest hardware, and is enough to learn how in-browser inference behaves before you scale up.\"}}\n  ]\n}\n<\/script>\n","protected":false},"excerpt":{"rendered":"<p>Ultimate Browser AI Models Directory Ultimate 2026 browser ai models directory with 85+ models, clear \ud83d\udfe2 offline browser capable highlighting, and new \u201cComputer Languages \/ Integration\u201d column showing exactly which languages &#038; frameworks (JavaScript, WebGPU, WASM, ONNX, TF.js etc.) can handle &#038; connect each model. Perfect reference for PrimeToolHub offline web tools. Picking a model &#8230; <a title=\"Ultimate Browser AI Models Directory 2026 | Offline WebGPU WASM Models\" class=\"read-more\" href=\"https:\/\/schoolict.net\/tools\/browser-ai-models-directory\/\" aria-label=\"Read more about Ultimate Browser AI Models Directory 2026 | Offline WebGPU WASM Models\">Read more<\/a><\/p>\n","protected":false},"author":1,"featured_media":2728,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"footnotes":""},"class_list":["post-2717","page","type-page","status-publish","has-post-thumbnail"],"_links":{"self":[{"href":"https:\/\/schoolict.net\/tools\/wp-json\/wp\/v2\/pages\/2717","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/schoolict.net\/tools\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/schoolict.net\/tools\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/schoolict.net\/tools\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/schoolict.net\/tools\/wp-json\/wp\/v2\/comments?post=2717"}],"version-history":[{"count":1,"href":"https:\/\/schoolict.net\/tools\/wp-json\/wp\/v2\/pages\/2717\/revisions"}],"predecessor-version":[{"id":7245,"href":"https:\/\/schoolict.net\/tools\/wp-json\/wp\/v2\/pages\/2717\/revisions\/7245"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/schoolict.net\/tools\/wp-json\/wp\/v2\/media\/2728"}],"wp:attachment":[{"href":"https:\/\/schoolict.net\/tools\/wp-json\/wp\/v2\/media?parent=2717"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}