Files
purpleandClaude Opus 5.5 b86747baee 3.0.8
- Web search with five services in turn, no key needed, pages read like a reader view
- A long video renders in the background, videos made in the chat play there, Wan 2.1 saves mp4
- A graphics card per task with two cards, AMD on Windows gets the current ROCm build, Intel Arc its own
- The LU Engine plans memory from the model file, says what a long context costs, keeps one copy of the model
- Settings from the chat with a confirmation card, Storage puts models, gallery and runtime in one folder
- Remove refusals from a local model in the Model Manager
- Native HiRes in Edit, LoRA stack with search and strength from -20 to 20
- Ask in Code asks before every external tool, REA shows only in Agent and Code
- Your phone on this computer (Remote), paired by QR code
- Wan 2.2 14B runs from Animate and Video, Train new on a Mac shows the trigger word field
- Temperature is held to 1.5, memories can be deleted, the onboarding runs once
- GitHub issues 187, 188, 190, 191, 193 and 194

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-27 02:52:46 -07:00

817 lines
85 KiB
HTML

<!DOCTYPE html>
<html lang="en" data-theme="dark">
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>Uncensored AI Studio: Free, Open Source Chat, Image, Video</title>
<meta name="description" content="Uncensored AI for chat, images, video and a coding agent. Free and open source on GitHub, on your own PC, or online without a GPU. Local chats stay on your PC.">
<meta name="author" content="PurpleDoubleD">
<meta name="application-name" content="Locally Uncensored">
<meta name="apple-mobile-web-app-title" content="Locally Uncensored">
<meta name="format-detection" content="telephone=no">
<meta name="robots" content="index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1">
<meta property="og:title" content="Uncensored AI Studio: Free, Open Source Chat, Image, Video">
<meta property="og:description" content="Uncensored AI studio for Windows, Linux and Mac: chat with Qwen, DeepSeek and Llama, FLUX images and Wan video via ComfyUI, plus a built-in coding agent in one installer. Free and open source on GitHub, or online in a browser when you have no GPU. Auto-detects Ollama, LM Studio, vLLM, llama.cpp, KoboldCpp. Local chats stay on your PC. AGPL-3.0.">
<meta property="og:image" content="https://raw.githubusercontent.com/PurpleDoubleD/locally-uncensored/master/docs/social-preview.png">
<meta property="og:image:width" content="2752">
<meta property="og:image:height" content="1536">
<meta property="og:image:alt" content="Locally Uncensored desktop app, private local AI for chat, code, image, and video generation">
<meta property="og:url" content="https://locallyuncensored.com/">
<meta property="og:type" content="website">
<meta property="og:site_name" content="Locally Uncensored">
<meta property="og:locale" content="en_US">
<meta name="twitter:card" content="summary_large_image">
<meta name="twitter:title" content="Uncensored AI Studio: Free, Open Source Chat, Image, Video">
<meta name="twitter:description" content="Uncensored AI for chat, code, FLUX images and Wan video on your own PC, or online without a GPU. Free and open source on GitHub. Local chats stay on your PC.">
<meta name="twitter:image" content="https://raw.githubusercontent.com/PurpleDoubleD/locally-uncensored/master/docs/social-preview.png">
<meta name="twitter:image:alt" content="Locally Uncensored desktop app, chat, code, image, video, all local">
<link rel="canonical" href="https://locallyuncensored.com/">
<meta name="theme-color" content="#0a0a0a">
<link rel="icon" type="image/svg+xml" href="/favicon.svg">
<link rel="icon" type="image/png" href="/favicon.png">
<link rel="apple-touch-icon" href="/favicon.png">
<link rel="manifest" href="/manifest.json">
<script type="application/ld+json">
{"@context":"https://schema.org","@type":"SoftwareApplication","@id":"https://locallyuncensored.com/#software","name":"Locally Uncensored","alternateName":["LU","Locally Uncensored Desktop App"],"publisher":{"@id":"https://locallyuncensored.com/#organization"},"applicationCategory":"DeveloperApplication","applicationSubCategory":"Artificial Intelligence","operatingSystem":"Windows 10, Windows 11, Linux, macOS 14 or newer (Apple Silicon)","description":"Uncensored AI studio, plug and play: chat with 20+ providers, generate images and video via ComfyUI, and code with a built-in AI agent, all on your own hardware. One installer, no command line, no Docker. Auto-detects 12 local backends. No analytics kit and no crash reports. Cloud only when you switch it on.","url":"https://locallyuncensored.com/","downloadUrl":"https://github.com/PurpleDoubleD/locally-uncensored/releases/latest","softwareVersion":"3.0.7","license":"https://www.gnu.org/licenses/agpl-3.0.html","author":{"@type":"Person","name":"PurpleDoubleD","url":"https://github.com/PurpleDoubleD"},"offers":{"@type":"Offer","price":"0","priceCurrency":"USD","availability":"https://schema.org/InStock"},"featureList":["Plug & Play Setup with 12 Local Backend Auto-Detection (Ollama, LM Studio, vLLM, KoboldCpp, Jan, llama.cpp, LocalAI, GPT4All, TabbyAPI, Aphrodite, SGLang, TGI)","20+ Provider Presets (local + cloud)","Qwen 3.8 27B in the Model Manager (vision via a separate mmproj file, official, abliterated and uncensored builds, plus a 9B distill)","GPT-OSS Compatible (120B, 20B via Ollama)","GLM-4.7 Flash Integration","DeepSeek R1 and DeepSeek V3 Support","Llama 4 and Llama 3.3 Ready","Gemma 4 E4B and 27B (Native Vision)","Mistral Small 3 and Phi 4 Support","Qwen 3 Coder and Qwen 2.5 Coder","Abliterated Model Variants","Coding Agent with Live Streaming and Apply-Patch","Claude Code CLI Integration","Agent Mode with 14 Tools, Parallel Execution, MCP, Sub-Agent Delegation","Remote Access from Phone via LAN or Cloudflare Tunnel (6-digit Passcode, QR Setup)","Mobile Web App (Chat, Coding Agent, Tools, Plugins)","Image Generation via ComfyUI (FLUX 2 Klein, FLUX.1, Juggernaut XL, Z-Image Turbo, ERNIE-Image, SDXL, SD 3.5)","Video Generation (Wan 2.1, Wan 2.2, HunyuanVideo 1.5, LTX 2.3, FramePack F1, AnimateDiff, Mochi, Cosmos)","Image-to-Image with Denoise Control","Image-to-Video (FramePack F1, SVD)","Local Create Lanes for Talking Character, Music, Extend Video and Motion Control (own GPU, core ComfyUI nodes)","Per-Lane Frame, Resolution and Step Controls","Surgical file_edit Tool in the Coding Agent","Remote Access to Non-Ollama Backends (LM Studio, Lemonade, llama.cpp) with Streaming, Tool Calls, Vision and Reasoning","Granular Permission System (7 Tool Categories)","File Upload with Vision Support","Thinking Mode with Universal Tag Stripping","A/B Model Compare","Local Benchmark","Memory System and Document RAG","Voice: a spoken conversation with your chat model, 15 voices, runs locally","Dictation and read aloud (Whisper, Piper)","Computer control: a model uses mouse and keyboard in apps you approve, off by default","Auto-Update over Signed NSIS Channel","75+ One-Click Model Downloads","Hardware-Aware Model Recommendations","Local models run offline. No analytics kit, no crash reports. Cloud is opt in.","AGPL-3.0 Open Source"],"screenshot":"https://raw.githubusercontent.com/PurpleDoubleD/locally-uncensored/master/docs/social-preview.png","softwareRequirements":"Any OpenAI-compatible local backend (Ollama, LM Studio, vLLM, KoboldCpp, etc.) or cloud API key","memoryRequirements":"8 GB RAM minimum; 8-16 GB VRAM for image/video generation","storageRequirements":"6 GB for default model"}
</script>
<script type="application/ld+json">
{"@context":"https://schema.org","@type":"Organization","@id":"https://locallyuncensored.com/#organization","name":"PurpleDoubleD","url":"https://github.com/PurpleDoubleD","logo":"https://locallyuncensored.com/logos/LU-monogram-bw-transparent.png","founder":{"@id":"https://locallyuncensored.com/#founder"},"sameAs":["https://github.com/PurpleDoubleD","https://github.com/PurpleDoubleD/locally-uncensored","https://reddit.com/user/GroundbreakingMall54","https://locallyuncensored.com"]}
</script>
<script type="application/ld+json">
{"@context":"https://schema.org","@type":"Person","@id":"https://locallyuncensored.com/#founder","name":"David Drobnitza","alternateName":"PurpleDoubleD","url":"https://locallyuncensored.com/","jobTitle":"Founder","worksFor":{"@id":"https://locallyuncensored.com/#organization"},"sameAs":["https://www.linkedin.com/in/david-drobnitza-locallyuncensored/","https://github.com/PurpleDoubleD"]}
</script>
<script type="application/ld+json">
{"@context":"https://schema.org","@type":"WebSite","@id":"https://locallyuncensored.com/#website","name":"Locally Uncensored","alternateName":"LU","url":"https://locallyuncensored.com/","publisher":{"@id":"https://locallyuncensored.com/#organization"},"description":"Uncensored AI studio: chat, image and video generation, and a coding agent in one app. Runs on your own PC or Mac, or online in a browser. Open source under AGPL-3.0."}
</script>
<script type="application/ld+json">
{"@context":"https://schema.org","@type":"HowTo","name":"How to install and run Locally Uncensored","description":"Install the desktop app and start running local AI in under five minutes.","totalTime":"PT5M","tool":[{"@type":"HowToTool","name":"Windows 10/11, Linux, or macOS 14 or newer on Apple Silicon"},{"@type":"HowToTool","name":"NVIDIA or AMD GPU (recommended for image and video generation)"}],"step":[{"@type":"HowToStep","position":1,"name":"Download and install","text":"Download the installer for Windows, Linux or macOS from GitHub Releases. One-click install, no Docker, no terminal, no config files.","url":"https://github.com/PurpleDoubleD/locally-uncensored/releases/latest"},{"@type":"HowToStep","position":2,"name":"Auto-detect local backends","text":"On first launch, the setup wizard scans for 12 local AI backends (Ollama, LM Studio, vLLM, KoboldCpp, Jan, and others) and offers one-click install links if none are running."},{"@type":"HowToStep","position":3,"name":"Chat, code, create","text":"Pick a model from the Model Manager and start chatting. Switch to the Coding Agent for coding tasks, or open Create to generate images and video. Add cloud providers (OpenAI, Anthropic, OpenRouter) in Settings if you want to mix local and cloud."}]}
</script>
<script type="application/ld+json">
{"@context":"https://schema.org","@type":"ItemList","name":"AI Models supported in Locally Uncensored","description":"Seventy-five-plus AI models for chat, coding, image generation, and video generation, all runnable locally via one-click download.","numberOfItems":22,"itemListElement":[{"@type":"ListItem","position":1,"name":"Qwen 3.8 27B","description":"Dense 27B in the Model Manager since v2.6.6. Vision works through a separate mmproj file the download writes next to the model."},{"@type":"ListItem","position":2,"name":"Qwen 3.8 9B distill","description":"The small checkpoint from the same wave, for cards that cannot hold the 27B. Text only, its repo ships no image projector."},{"@type":"ListItem","position":3,"name":"Qwen 3 Coder","description":"Native tool calling, specialised for coding."},{"@type":"ListItem","position":4,"name":"GPT-OSS-120B","description":"OpenAI open-weight release, runs via Ollama."},{"@type":"ListItem","position":5,"name":"GPT-OSS-20B","description":"Smaller GPT-OSS variant for faster inference."},{"@type":"ListItem","position":6,"name":"GLM-4.7 Flash","description":"Zhipu AI fast chat model."},{"@type":"ListItem","position":7,"name":"DeepSeek R1","description":"Reasoning model with thinking chains."},{"@type":"ListItem","position":8,"name":"Llama 4","description":"Meta flagship with vision and long context."},{"@type":"ListItem","position":9,"name":"Llama 3.3","description":"Stable instruction-tuned Meta model."},{"@type":"ListItem","position":10,"name":"Gemma 4","description":"Google flagship. Native tools, vision. E4B (4 GB) and 27B (16 GB)."},{"@type":"ListItem","position":11,"name":"Mistral Small 3","description":"Compact performant Mistral model."},{"@type":"ListItem","position":12,"name":"Phi 4","description":"Microsoft Phi series."},{"@type":"ListItem","position":13,"name":"FLUX 2 Klein","description":"Next-gen FLUX image model. 8-10 GB VRAM. T2I and I2I."},{"@type":"ListItem","position":14,"name":"Juggernaut XL","description":"Popular SDXL finetune. Photorealism. 8-10 GB VRAM."},{"@type":"ListItem","position":15,"name":"Z-Image Turbo","description":"Uncensored image generation. 10-16 GB VRAM."},{"@type":"ListItem","position":16,"name":"ERNIE-Image","description":"Baidu image generation (Turbo 8-step and Base 50-step)."},{"@type":"ListItem","position":17,"name":"SDXL and SD 3.5","description":"Stable Diffusion baseline models."},{"@type":"ListItem","position":18,"name":"Wan 2.1","description":"Text-to-video. Wan 1.3B (8 GB) or 14B (12+ GB)."},{"@type":"ListItem","position":19,"name":"HunyuanVideo 1.5","description":"Text-to-video model for consistency."},{"@type":"ListItem","position":20,"name":"LTX 2.3","description":"Lightricks LTX-Video. Fast text-to-video on modest hardware."},{"@type":"ListItem","position":21,"name":"FramePack F1","description":"Image-to-video on just 6 GB VRAM. Next-frame prediction."},{"@type":"ListItem","position":22,"name":"AnimateDiff","description":"Animation for SD-based models."}]}
</script>
<script type="application/ld+json">
{"@context":"https://schema.org","@type":"FAQPage","mainEntity":[{"@type":"Question","name":"What is Locally Uncensored?","acceptedAnswer":{"@type":"Answer","text":"A free, open-source local AI studio for the desktop. Install it like any normal app and you're chatting, generating images, and making videos in minutes, with no command line and no Docker. It combines chat (20+ provider presets), a coding agent, image generation via ComfyUI (FLUX 2, Juggernaut XL, Z-Image, SDXL), and video generation (Wan 2.1, HunyuanVideo, LTX 2.3, FramePack F1) in one interface. AGPL-3.0 licensed."}},{"@type":"Question","name":"Does Locally Uncensored support Qwen 3.8?","acceptedAnswer":{"@type":"Answer","text":"Yes. Qwen 3.8 27B is in the Model Manager since v2.6.6, with working vision: the image tower ships as a separate mmproj file that the download writes next to the model. Official, huihui abliterated and OrcaRouter uncensored builds are there, plus a 9B distill for small cards and two Ollama tags."}},{"@type":"Question","name":"Does it support GPT-OSS?","acceptedAnswer":{"@type":"Answer","text":"Yes. GPT-OSS-120B and GPT-OSS-20B run via Ollama, which is one of the twelve auto-detected local backends."}},{"@type":"Question","name":"Does it support GLM-4.7?","acceptedAnswer":{"@type":"Answer","text":"Yes. GLM-4.7 Flash is supported through Ollama and other compatible backends. Eleven variants are available for one-click download."}},{"@type":"Question","name":"Does it support DeepSeek R1 and Llama 4?","acceptedAnswer":{"@type":"Answer","text":"Yes. DeepSeek R1, DeepSeek V3, Llama 4, Llama 3.3, Gemma 4, Mistral Small 3, Phi 4 and more are all supported through any of the twelve auto-detected backends."}},{"@type":"Question","name":"Is it really free and offline?","acceptedAnswer":{"@type":"Answer","text":"Yes. After initial setup and model download, no internet is needed for the local stack. Local use needs no account and has no usage limit. The app has no analytics kit and sends no crash reports. On its own it checks GitHub for updates and counts presses of the Cloud switch. Cloud providers (OpenAI, Anthropic, OpenRouter, Groq) are optional."}},{"@type":"Question","name":"How is this different from Open WebUI, LM Studio, Jan or Msty?","acceptedAnswer":{"@type":"Answer","text":"Those tools handle text chat. Locally Uncensored adds a coding agent with 14 MCP tools, ComfyUI image generation (FLUX, Juggernaut, Z-Image, SDXL), video generation (Wan 2.1, HunyuanVideo, LTX 2.3, FramePack), mobile remote access via LAN or Cloudflare Tunnel, A/B model compare, local benchmarking, granular permissions, file upload with vision, and thinking mode. All in one app."}},{"@type":"Question","name":"What hardware do I need?","acceptedAnswer":{"@type":"Answer","text":"Text chat: 8 GB RAM. Image generation: an NVIDIA card, or an AMD card with ROCm on Linux, from 6 GB of VRAM with the starter checkpoint, 8 to 10 GB for FLUX. Video generation: from 6 GB for AnimateDiff and image to video with FramePack F1, from 8 GB for the Wan 2.1 starter, 12 GB and up for Wan 2.2 and HunyuanVideo. Windows 10/11 and Linux supported, and Apple Silicon Macs since version 3.0.6."}},{"@type":"Question","name":"What does uncensored mean?","acceptedAnswer":{"@type":"Answer","text":"Abliterated AI models with artificial refusal behaviour removed. The AI responds honestly without disclaimers. Combined with fully local execution, your conversations stay private."}},{"@type":"Question","name":"Can I use it as a ChatGPT or Claude alternative?","acceptedAnswer":{"@type":"Answer","text":"Yes. Locally Uncensored is a ChatGPT and Claude alternative that runs on your own hardware. Use Qwen 3.8, GPT-OSS, GLM-4.7, DeepSeek R1, Llama 4 or Gemma 4 instead, or add cloud providers if you want them alongside the local stack."}},{"@type":"Question","name":"Does remote access leak data?","acceptedAnswer":{"@type":"Answer","text":"Remote access is off until you start it, sits behind a 6-digit passcode, and shows when a device is connected. On your own network the chat goes straight from phone to PC. Over the internet Cloudflare relays it: the traffic is encrypted to and from Cloudflare, not end to end, so Cloudflare could read it."}},{"@type":"Question","name":"Can I run it on macOS?","acceptedAnswer":{"@type":"Answer","text":"Yes, since version 3.0.6, in beta. The Mac build is for Apple Silicon and needs macOS 14 or newer. Intel Macs are not supported. The .dmg is signed and notarized. Chat runs on local models through the bundled server with Metal, and Create makes images and video on the Mac itself, without ComfyUI. We tested it on an M5 Pro with 48 GB. Tell us how it runs on yours. Voice has been tried by hand in the app on a Mac and works. Computer control is in the Mac app but not fully tested yet on macOS. Remote Access and the automatic update are not confirmed on macOS. On any Mac you can also use the hosted studio in a browser at lu-labs.ai."}},{"@type":"Question","name":"Is there a mobile app?","acceptedAnswer":{"@type":"Answer","text":"Yes. The desktop app hosts a full mobile web app over LAN or Cloudflare Tunnel. Chat, Coding Agent, 14 agent tools, plugins, personas, all from your phone."}},{"@type":"Question","name":"Can I talk to it?","acceptedAnswer":{"@type":"Answer","text":"Yes. Voice is a spoken conversation with the chat model you picked. Speech recognition and the voice run on your own machine, in English, with fifteen voices. Voice is in beta: our check in the real app ran on Windows, and your computer may still show rough edges."}},{"@type":"Question","name":"Can it control my computer?","acceptedAnswer":{"@type":"Answer","text":"Only if you switch it on. With computer control a model uses the mouse and keyboard in apps you approve. It asks before it touches an app and, by default, before every action, and it never touches terminals, password managers or password fields. You can pause or stop it at any time. Computer control is in beta: it is not proven in every app yet, so watch what it does and stop it when it goes wrong."}}]}
</script>
<link rel="preconnect" href="https://fonts.googleapis.com">
<link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
<link href="https://fonts.googleapis.com/css2?family=Inter:wght@200;300;400;500;600;700;800&family=JetBrains+Mono:wght@400;500&family=Space+Grotesk:wght@500;600;700&display=swap" rel="stylesheet">
<link rel="stylesheet" href="/assets/lu.css">
<script defer src="/assets/lu.js"></script>
<script data-goatcounter="https://purple47.goatcounter.com/count" async src="//gc.zgo.at/count.js"></script>
<link rel="stylesheet" href="/assets/home.css">
<script defer src="/assets/home-motion.js"></script>
</head>
<body class="hp">
<script>document.documentElement.classList.add("js")</script>
<script>
// The download buttons follow the visitor's system. Every link stands in the
// markup, and assets/home.css shows the set that html[data-os] names. Without
// this script the page keeps the Windows set and offers the others as links.
(function () {
var root = document.documentElement, nav = navigator, hint = nav.userAgentData;
var seen = ((hint && hint.platform) || nav.platform || '') + ' ' + (nav.userAgent || '');
var os = 'win';
// A phone or tablet cannot run the desktop app. It gets LU Cloud first and
// a way to send the download link to a computer (html[data-device]).
var phone = (hint && hint.mobile) || /android|iphone|ipad|ipod/i.test(seen) || (/mac/i.test(seen) && nav.maxTouchPoints > 1);
if (phone) root.setAttribute('data-device', 'phone');
if ((hint && hint.mobile) || /android|iphone|ipad|ipod|cros/i.test(seen)) os = 'win';
else if (/mac/i.test(seen)) os = nav.maxTouchPoints > 1 ? 'win' : 'mac';
else if (/linux|x11/i.test(seen)) os = 'linux';
function choose(next) {
root.setAttribute('data-os', next);
var picks = document.querySelectorAll('[data-dl-os]');
for (var i = 0; i < picks.length; i++) {
if (picks[i].getAttribute('data-dl-os') === next) picks[i].setAttribute('aria-current', 'true');
else picks[i].removeAttribute('aria-current');
}
}
choose(os);
document.addEventListener('DOMContentLoaded', function () { choose(root.getAttribute('data-os')); });
document.addEventListener('click', function (event) {
var pick = event.target.closest ? event.target.closest('[data-dl-os]') : null;
if (!pick) return;
event.preventDefault();
choose(pick.getAttribute('data-dl-os'));
});
})();
</script>
<div class="lu-bg hp-bg" aria-hidden="true"><i class="hp-glow g1"></i><i class="hp-glow g2"></i><i class="hp-glow g3"></i></div>
<div class="lu-atmosphere" aria-hidden="true"></div>
<nav class="lu-nav">
<a class="lu-nav-brand" href="#" aria-label="Locally Uncensored">
<svg class="hp-mark" viewBox="176 176 672 672" aria-hidden="true"><defs><linearGradient id="hp-mark-g" x1="0" y1="0" x2="1" y2="1"><stop offset="0" stop-color="#a78bfa"/><stop offset="1" stop-color="#8b5cf6"/></linearGradient></defs><rect x="193" y="193" width="638" height="638" rx="138" ry="138" fill="none" stroke="url(#hp-mark-g)" stroke-width="44"/><path fill="url(#hp-mark-g)" d="M302,342 H377 V612 H433.6 A162,162 0 0,0 500.5,680 H302 Z"/><path fill="url(#hp-mark-g)" d="M434,342 H508 V545 A69,69 0 0,0 646,545 V342 H722 V537.5 A144,144 0 0,1 434,537.5 Z"/></svg>
<span class="lu-nav-word">Locally <span class="accent">Uncensored</span></span>
</a>
<div class="lu-nav-links" id="lu-nav-links">
<a class="lu-nav-link" href="#things">What</a>
<a class="lu-nav-link" href="#models">Models</a>
<a class="lu-nav-link" href="/pricing/">Pricing</a>
<a class="lu-nav-link" href="/cloud/">Cloud</a>
<a class="lu-nav-link" href="#how">Setup</a>
<a class="lu-nav-link" href="#blog">Blog</a>
<a class="lu-nav-link" href="#faq">FAQ</a>
</div>
<div class="lu-nav-right">
<a class="lu-nav-icon" href="https://github.com/PurpleDoubleD/locally-uncensored" title="GitHub" aria-label="GitHub"><svg width="17" height="17" viewBox="0 0 24 24" fill="currentColor" aria-hidden="true"><path d="M12 .5C5.37.5 0 5.87 0 12.5c0 5.3 3.44 9.8 8.21 11.39.6.11.82-.26.82-.58 0-.29-.01-1.04-.02-2.05-3.34.73-4.04-1.61-4.04-1.61-.55-1.39-1.33-1.76-1.33-1.76-1.09-.75.08-.73.08-.73 1.2.08 1.84 1.24 1.84 1.24 1.07 1.83 2.81 1.3 3.5.99.11-.78.42-1.3.76-1.6-2.67-.3-5.47-1.33-5.47-5.93 0-1.31.47-2.38 1.24-3.22-.13-.3-.54-1.52.11-3.18 0 0 1.01-.32 3.3 1.23a11.5 11.5 0 0 1 6.01 0c2.29-1.55 3.3-1.23 3.3-1.23.65 1.66.24 2.88.12 3.18.77.84 1.23 1.91 1.23 3.22 0 4.61-2.8 5.63-5.48 5.92.43.37.81 1.1.81 2.22 0 1.6-.01 2.9-.01 3.29 0 .32.21.7.82.58A12.01 12.01 0 0 0 24 12.5C24 5.87 18.63.5 12 .5Z"/></svg><span class="lu-nav-badge">2000+&#8202;⭐</span></a>
<a class="lu-nav-icon discord" href="https://locallyuncensored.com/discord" target="_blank" rel="noopener" title="Discord" aria-label="Discord"><svg width="18" height="18" viewBox="0 0 24 24" fill="currentColor" aria-hidden="true"><path d="M20.32 4.37A19.8 19.8 0 0 0 15.45 3c-.2.36-.44.85-.6 1.23a18.3 18.3 0 0 0-5.7 0c-.16-.38-.4-.87-.6-1.23a19.7 19.7 0 0 0-4.87 1.37C.55 8.9-.32 13.3.11 17.63a19.9 19.9 0 0 0 6 3.03c.48-.66.9-1.36 1.27-2.1-.7-.26-1.36-.58-1.99-.96.17-.12.33-.25.49-.38a14.2 14.2 0 0 0 12.24 0c.16.13.32.26.49.38-.63.38-1.3.7-2 .96.37.74.8 1.44 1.27 2.1a19.8 19.8 0 0 0 6-3.03c.5-5.03-.85-9.39-3.55-13.26ZM8.02 15c-1.18 0-2.15-1.09-2.15-2.42 0-1.34.95-2.43 2.15-2.43 1.2 0 2.17 1.09 2.15 2.43 0 1.33-.95 2.42-2.15 2.42Zm7.96 0c-1.18 0-2.15-1.09-2.15-2.42 0-1.34.95-2.43 2.15-2.43 1.2 0 2.17 1.09 2.15 2.43 0 1.33-.94 2.42-2.15 2.42Z"/></svg><span class="lu-nav-badge">1200+&#8202;<span style="font-size:0.715em">🟢</span></span></a>
<a class="he-cta" href="/pricing/" aria-label="LU Cloud plans and what they include"><svg viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2.2" stroke-linecap="round" stroke-linejoin="round" aria-hidden="true"><path d="M17.5 19H9a7 7 0 1 1 6.71-9h1.79a4.5 4.5 0 1 1 0 9Z"/></svg>Cloud</a>
<span class="lu-nav-sep"></span>
<div class="switch" id="theme-switch" role="button" aria-label="Toggle theme"></div>
<a class="lu-nav-dl" data-dl-for="win" href="https://github.com/PurpleDoubleD/locally-uncensored/releases/latest/download/Locally.Uncensored_3.0.7_x64-setup.exe" download><svg width="14" height="14" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2.2" stroke-linecap="round" stroke-linejoin="round" aria-hidden="true"><path d="M12 3v12m0 0 4-4m-4 4-4-4"/><path d="M4 17v2a2 2 0 0 0 2 2h12a2 2 0 0 0 2-2v-2"/></svg>Download</a>
<a class="lu-nav-dl" data-dl-for="mac" href="https://github.com/PurpleDoubleD/locally-uncensored/releases/latest/download/LU_3.0.7_aarch64.dmg" download><svg width="14" height="14" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2.2" stroke-linecap="round" stroke-linejoin="round" aria-hidden="true"><path d="M12 3v12m0 0 4-4m-4 4-4-4"/><path d="M4 17v2a2 2 0 0 0 2 2h12a2 2 0 0 0 2-2v-2"/></svg>Download</a>
<a class="lu-nav-dl" data-dl-for="linux" href="https://github.com/PurpleDoubleD/locally-uncensored/releases/latest/download/Locally.Uncensored_3.0.7_amd64.AppImage" download><svg width="14" height="14" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2.2" stroke-linecap="round" stroke-linejoin="round" aria-hidden="true"><path d="M12 3v12m0 0 4-4m-4 4-4-4"/><path d="M4 17v2a2 2 0 0 0 2 2h12a2 2 0 0 0 2-2v-2"/></svg>Download</a>
<button type="button" class="lu-nav-toggle" aria-label="Menu">
<svg width="20" height="20" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" aria-hidden="true"><path d="M4 7h16M4 12h16M4 17h16"/></svg>
</button>
</div>
</nav>
<!-- ── HERO: headline, the download card, then the studio as a live reel ── -->
<header class="lu-hero hx">
<div class="hx-grid" aria-hidden="true"></div>
<a class="hx-pill" href="/pricing/"><i class="hx-live" aria-hidden="true"></i><b>LU Cloud is live</b><span class="hx-pill-sub">No GPU? Run it in your browser</span><svg viewBox="0 0 24 24" aria-hidden="true"><path d="M5 12h14M13 6l6 6-6 6"/></svg></a>
<h1 class="mast-title hx-title"><span class="accent hx-rx">Uncensored&nbsp;AI</span>,<br class="hx-br"> on your PC or online</h1>
<p class="mast-kicker hx-kicker"><span>Locally Uncensored, the plug-and-play AI studio</span><span class="k-dot">·</span><span>Chat · Code · Image · Video · Remote</span></p>
<p class="section-lead mast-lede hx-lede">An uncensored AI answers the question instead of refusing it. You can run one free on your own PC, or open it in a browser when the GPU is not there. Read <a href="/uncensored-ai/">what uncensored AI means here</a>.</p>
<div class="hx-dl">
<div class="hx-phone">
<a class="dl-btn" href="https://lu-labs.ai/app?demo=1&amp;src=luc-home-phone"><svg width="14" height="14" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2.4" stroke-linecap="round" stroke-linejoin="round" aria-hidden="true"><path d="M5 12h14M13 6l6 6-6 6"/></svg>Try LU Cloud in this browser</a>
<p class="hx-phone-sub">The demo needs no account. Credit packs from 15&nbsp;EUR, plans from 19&nbsp;EUR a month.</p>
<button type="button" class="hx-phone-share" data-share-desktop><svg width="14" height="14" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2.2" stroke-linecap="round" stroke-linejoin="round" aria-hidden="true"><path d="M4 12v7a2 2 0 0 0 2 2h12a2 2 0 0 0 2-2v-7"/><path d="M16 6l-4-4-4 4"/><path d="M12 2v13"/></svg><span>Desktop app for Windows, Mac and Linux: send the link to yourself</span></button>
<p class="hx-phone-done" data-share-status role="status" aria-live="polite"></p>
</div>
<div class="dl-pick">
<div class="dl-card dl-card--xl">
<span class="dl-win" aria-hidden="true"><svg data-dl-for="win" viewBox="0 0 24 24"><path d="M3 5.5 10.5 4.4v7.1H3zM11.5 4.3 21 3v8.5h-9.5zM3 12.5h7.5v7.1L3 18.5zM11.5 12.5H21V21l-9.5-1.3z"/></svg><svg data-dl-for="mac" viewBox="0 0 24 24"><path d="M12.152 6.896c-.948 0-2.415-1.078-3.96-1.04-2.04.027-3.91 1.183-4.961 3.014-2.117 3.675-.546 9.103 1.519 12.09 1.013 1.454 2.208 3.09 3.792 3.039 1.52-.065 2.09-.987 3.935-.987 1.831 0 2.35.987 3.96.948 1.637-.026 2.676-1.48 3.676-2.948 1.156-1.688 1.636-3.325 1.662-3.415-.039-.013-3.182-1.221-3.22-4.857-.026-3.04 2.48-4.494 2.597-4.559-1.429-2.09-3.623-2.324-4.39-2.376-2-.156-3.675 1.09-4.61 1.09zM15.53 3.83c.843-1.012 1.4-2.427 1.245-3.83-1.207.052-2.662.805-3.532 1.818-.78.896-1.454 2.338-1.273 3.714 1.338.104 2.715-.688 3.559-1.701"/></svg><svg data-dl-for="linux" viewBox="3 1.5 18 21"><path fill-rule="evenodd" d="M12 2c-2.2 0-3.6 1.8-3.6 4.2 0 1.5.2 2.4-.6 3.8C6.6 12.2 5 14.6 5 17.4c0 1.2.4 2.2 1 3-.6.3-1 .8-1 1.1 0 .4.6.5 1.6.5h10.8c1 0 1.6-.1 1.6-.5 0-.3-.4-.8-1-1.1.6-.8 1-1.8 1-3 0-2.8-1.6-5.2-2.8-7.4-.8-1.4-.6-2.3-.6-3.8C15.6 3.8 14.2 2 12 2zm0 8.5c1.8 0 3.6 3 3.6 6 0 2-1.4 3.2-3.6 3.2s-3.6-1.2-3.6-3.2c0-3 1.8-6 3.6-6zM10.6 5.5a.7.7 0 1 0 0 1.4.7.7 0 0 0 0-1.4zm2.8 0a.7.7 0 1 0 0 1.4.7.7 0 0 0 0-1.4z"/></svg></span>
<div class="dl-info">
<h3>Locally Uncensored <em>&middot; 3.0.7</em></h3>
<p>v3.0.7 &middot; <span data-dl-for="win">WINDOWS</span> <span data-dl-for="mac">MACOS (BETA) &middot; APPLE SILICON</span> <span data-dl-for="linux">LINUX</span> &middot; <span class="dl-nb">170,000+ DOWNLOADS</span></p>
</div>
<a class="dl-btn" data-dl-for="win" href="https://github.com/PurpleDoubleD/locally-uncensored/releases/latest/download/Locally.Uncensored_3.0.7_x64-setup.exe" download><svg width="14" height="14" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2.4" stroke-linecap="round" stroke-linejoin="round" aria-hidden="true"><path d="M21 15v4a2 2 0 0 1-2 2H5a2 2 0 0 1-2-2v-4"/><polyline points="7 10 12 15 17 10"/><line x1="12" y1="15" x2="12" y2="3"/></svg>Download for Windows (.exe)</a>
<a class="dl-btn" data-dl-for="mac" href="https://github.com/PurpleDoubleD/locally-uncensored/releases/latest/download/LU_3.0.7_aarch64.dmg" download><svg width="14" height="14" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2.4" stroke-linecap="round" stroke-linejoin="round" aria-hidden="true"><path d="M21 15v4a2 2 0 0 1-2 2H5a2 2 0 0 1-2-2v-4"/><polyline points="7 10 12 15 17 10"/><line x1="12" y1="15" x2="12" y2="3"/></svg>Download for Mac (.dmg)</a>
<a class="dl-btn" data-dl-for="linux" href="https://github.com/PurpleDoubleD/locally-uncensored/releases/latest/download/Locally.Uncensored_3.0.7_amd64.AppImage" download><svg width="14" height="14" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2.4" stroke-linecap="round" stroke-linejoin="round" aria-hidden="true"><path d="M21 15v4a2 2 0 0 1-2 2H5a2 2 0 0 1-2-2v-4"/><polyline points="7 10 12 15 17 10"/><line x1="12" y1="15" x2="12" y2="3"/></svg>Download for Linux (AppImage)</a>
</div>
<p class="dl-android"><a href="https://github.com/PurpleDoubleD/locally-uncensored/releases/latest/download/LU-3.0.7-Android.apk" download>Android phone? Download the APK</a> &middot; Android 7.0 or newer, 64-bit, 27 MB &middot; <a href="/android/">how to install</a></p>
<div class="dl-os" role="group" aria-label="Choose your system">
<a data-dl-os="win" href="https://github.com/PurpleDoubleD/locally-uncensored/releases/latest/download/Locally.Uncensored_3.0.7_x64-setup.exe" download aria-current="true">Windows</a>
<a data-dl-os="mac" href="https://github.com/PurpleDoubleD/locally-uncensored/releases/latest/download/LU_3.0.7_aarch64.dmg" download>macOS</a>
<a data-dl-os="linux" href="https://github.com/PurpleDoubleD/locally-uncensored/releases/latest/download/Locally.Uncensored_3.0.7_amd64.AppImage" download>Linux</a>
</div>
<p class="dl-note">
<span data-dl-note="win">Windows 10 and 11, x64.</span>
<span data-dl-note="mac">macOS is in beta: Apple Silicon, macOS 14 or newer. Intel Macs are not supported. Details on the <a href="/mac/">Mac page</a>.</span>
<span data-dl-note="linux">Linux x64: the AppImage runs on any distribution, or take the <a href="https://github.com/PurpleDoubleD/locally-uncensored/releases/latest/download/Locally.Uncensored_3.0.7_amd64.deb" download>.deb</a> or the <a href="https://github.com/PurpleDoubleD/locally-uncensored/releases/latest/download/Locally.Uncensored-3.0.7-1.x86_64.rpm" download>.rpm</a>.</span>
</p>
</div>
<div class="dl-card dl-card--xl dl-card--cloud">
<span class="dl-win dl-lu" aria-hidden="true"><svg viewBox="193 193 638 638"><rect x="193" y="193" width="638" height="638" rx="138" ry="138" fill="#8b5cf6"/><path fill="#fff" d="M302,342 H377 V612 H433.6 A162,162 0 0,0 500.5,680 H302 Z"/><path fill="#fff" d="M434,342 H508 V545 A69,69 0 0,0 646,545 V342 H722 V537.5 A144,144 0 0,1 434,537.5 Z"/></svg></span>
<div class="dl-info">
<h3>LU Cloud <em>&middot; in your browser</em></h3>
<p>NO GPU NEEDED &middot; MAC &middot; PHONE &middot; PACKS FROM 15 EUR</p>
</div>
<a class="dl-btn" href="https://lu-labs.ai/pricing?tab=credits&amp;src=luc-home-hero">
<svg width="14" height="14" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2.4" stroke-linecap="round" stroke-linejoin="round" aria-hidden="true"><path d="M5 12h14M13 6l6 6-6 6"/></svg>
See plans and packs
</a>
</div>
<p class="hx-alt">Free and open source. No GPU? <a href="/pricing/">Use LU Cloud</a>.</p>
</div>
<div class="dl-alts">
<a href="#whats-new">What's new</a>
<a href="https://locallyuncensored.com/discord" target="_blank" rel="noopener">Join Discord</a>
<a href="https://github.com/PurpleDoubleD/locally-uncensored/releases">All releases</a>
<a href="https://github.com/PurpleDoubleD/locally-uncensored">Source on GitHub</a>
<a href="https://github.com/PurpleDoubleD/locally-uncensored/discussions">Discussions</a>
<a href="https://github.com/PurpleDoubleD/locally-uncensored/blob/master/SECURITY.md#antivirus--browser-false-positives">Antivirus warning?</a>
</div>
</header>
<section class="hx-reel" id="content-start" aria-labelledby="reel-h">
<div class="hx-reel-in">
<div class="reel-bar">
<div class="reel-t">
<h2 class="reel-h" id="reel-h">Generate Anything. Locally. <em class="rx">Uncensored.</em></h2>
<span class="reel-k"><i aria-hidden="true"></i>Live demo · one character, start to finish</span>
</div>
<div class="mv-tabs" data-mv-for="mv-hero">
<button type="button" data-i="0"><span>01</span>Chat</button>
<button type="button" data-i="1"><span>02</span>Image</button>
<button type="button" data-i="2"><span>03</span>Video</button>
<button type="button" data-i="3"><span>04</span>Edit</button>
<button type="button" data-i="4"><span>05</span>Code</button>
</div>
</div>
<figure class="mv mv--hero" id="mv-hero" data-mv="hero">
<div class="mv-box" data-nosnippet style="--mv-poster:url(/assets/home/neon.webp)"><div class="mv-clip"><div class="mv-stage" aria-hidden="true" translate="no"></div></div></div>
<div class="mv-ctl"><button type="button" class="mv-pp" aria-label="Pause demo"></button><div class="mv-track"><i class="mv-fill"></i></div><span class="mv-tc">0:00</span></div>
<figcaption class="sr-only">A sixteen second demo of the app: a chat invents a character, Create draws her, turns the picture into a video clip and edits the scene, and the Coding Agent puts the picture on a web page.</figcaption>
</figure>
</div>
</section>
<!-- ── PROOF STRIP ─────────────────────────────────────────── -->
<div class="proof" aria-label="Locally Uncensored in numbers">
<div class="proof-row">
<span>170,000+ downloads</span><span>2,000+ GitHub stars</span><span>Free and open source, AGPL-3.0</span><span>Local chats stay on your PC</span><span>12 local backends detected for you</span><span>75+ one-click models</span><span>Windows, Linux and Mac</span><span>2,200+ automated checks</span>
<span aria-hidden="true">170,000+ downloads</span><span aria-hidden="true">2,000+ GitHub stars</span><span aria-hidden="true">Free and open source, AGPL-3.0</span><span aria-hidden="true">Local chats stay on your PC</span><span aria-hidden="true">12 local backends detected for you</span><span aria-hidden="true">75+ one-click models</span><span aria-hidden="true">Windows, Linux and Mac</span><span aria-hidden="true">2,200+ automated checks</span>
</div>
</div>
<!-- ── CLOUD (same catalog, in a browser) ───────────────── -->
<section class="models lu-cloud hp-cloud" id="cloud-online">
<div class="container">
<div class="hp-head">
<p class="eyebrow">No GPU, or the wrong one</p>
<h2>Prefer not to buy a GPU? <em class="rx">The same catalog runs online</em></h2>
<p class="section-lead">The hosted side opens in a browser, which is also how it works on a phone or on a computer too small for a model, and it is paid from credit packs that start at 15 EUR, never expire and need no subscription. LU Cloud runs on lu-labs.ai, our hosted service.</p>
</div>
<div class="hp-why">
<p class="hp-why-lead"><b>Already have a GPU?</b> A desktop card holds chat models up to about 30B. LU Cloud runs the ones no desktop card can hold, like <span data-model-id="moonshotai/Kimi-K3">Kimi K3</span> and <span data-model-id="Qwen/Qwen3.8-Max">Qwen 3.8 Max</span>, in the same app. Flip the Cloud switch when a job needs it, and switch back. Local stays free.</p>
<ul class="hp-why-row">
<li><b>Bigger models</b><span>Chat models too large for any desktop card.</span></li>
<li><b>Faster video</b><span>Clips render on our GPUs, so your own card stays free.</span></li>
<li><b>Phone and Intel Mac</b><span>The same studio in a browser, on any phone and on an Intel Mac.</span></li>
</ul>
</div>
<div class="cl-split">
<figure class="mv" id="mv-cloud" data-mv="cloud">
<div class="mv-top"><span class="mv-live"><i></i>Explainer</span><span class="mv-name">Where your prompt runs</span></div>
<div class="mv-box" data-nosnippet><div class="mv-clip"><div class="mv-stage" aria-hidden="true" translate="no"></div></div></div>
<div class="mv-ctl"><button type="button" class="mv-pp" aria-label="Pause demo"></button><div class="mv-track"><i class="mv-fill"></i></div><span class="mv-tc">0:00</span></div>
<figcaption class="sr-only">Explainer: in local mode the prompt stays inside your PC and runs on your graphics card. Switched to LU Cloud, the same app sends it to hosted GPUs, so no graphics card is needed.</figcaption>
</figure>
<div class="cl-stats">
<div class="cl-stat"><b data-count="48">47</b><span>chat models online, every one of them in every plan and every pack</span></div>
<div class="cl-stat"><b><span data-count="19">19</span><small>/44</small></b><span>measured chat models answer without refusing, marked in the picker</span></div>
<div class="cl-stat"><b data-count="7">7</b><span>open image models, plus <em>14</em> open video models</span></div>
<div class="cl-stat"><b><small>&euro;</small><span data-count="19">19</span></b><span>a month for Hosted, or credit packs from 15 EUR that never expire</span></div>
<p class="cl-foot">We add no moderation layer of our own to text. For cloud image and video, your LU Cloud account has a content policy setting: Strict, Standard, or off.</p>
</div>
</div>
<div class="lu-cloud-grid">
<article class="lu-cloud-card">
<h3>Uncensored AI chat online</h3>
<p>48 chat models, every one of them included in every plan and every pack. We add no moderation layer of our own to text.</p>
<a href="/uncensored-ai-chat-online/">Chat in the browser</a>
</article>
<article class="lu-cloud-card">
<h3>Uncensored AI image generator online</h3>
<p>Images from about 2 cents each on Hosted, or about 3 cents on a 15 EUR pack. The content policy setting in your account decides what the image models may produce.</p>
<a href="/uncensored-ai-image-generator-online/">See the image models</a>
</article>
<article class="lu-cloud-card">
<h3>Uncensored AI video generator online</h3>
<p>A five-second clip from about 21 cents on Hosted, or about 33 cents on a 15 EUR pack. You supply the starting frame; making that image is billed on its own.</p>
<a href="/uncensored-ai-video-generator-online/">See the video models</a>
</article>
<article class="lu-cloud-card">
<h3>Uncensored AI API</h3>
<p>The desktop app serves an OpenAI-compatible API for your local models on port 8129, free. For hosted models the endpoint is https://lu-labs.ai/api/inference/v1, with lu_ keys that spend credits.</p>
<a href="/uncensored-ai-api/">Read the API page</a>
</article>
</div>
<p class="blog-all">Seventeen open video models and nine open image models are listed with their prices on our <a href="/cloud/">cloud page</a>. The pricing page on lu-labs.ai lists only the general image and video models, so its counts are lower.</p>
<p class="blog-all"><a class="dl-btn" href="https://lu-labs.ai/pricing?tab=credits&amp;src=luc-home-nogpu"><svg width="14" height="14" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2.4" stroke-linecap="round" stroke-linejoin="round" aria-hidden="true"><path d="M5 12h14M13 6l6 6-6 6"/></svg> See plans and packs</a></p>
<div class="hp-facts">
<!-- LU-CLOUD-FACTS:START -->
<section><h2>Current Hosted offer</h2><p>Hosted includes 11 Flash models; Pro and Max include 12. In app sessions they cost no credits, up to a daily limit per account. GLM 5.3 Flash uses credits on Hosted and joins the daily allowance on Pro and Max. The allowance needs an active paid plan, and accounts without an active plan keep paying credits. API keys always use credits.</p><p>EUR 19/month, 900,000 shared monthly credits: up to 900 Neta Lumina Uncensored images or 90 five-second LTX 2.3 Uncensored clips. Every output spends the same wallet.</p><p>Each maximum spends the entire shared monthly wallet on one model. These are alternatives, not separate allowances. Input-image generation, character training, LoRA routes and additional operations cost extra. Uncensored video uses an input image and five-second clips. Failed jobs and changing catalog prices can affect results.</p><p><a href="/uncensored-ai-image-generator-online/">Image models and exact quantities</a> · <a href="/uncensored-ai-video-generator-online/">Video models and exact quantities</a> · <a href="/vs/venice/">Venice comparison</a> · <a href="/cloud-facts.json">Current facts JSON</a></p></section>
<!-- LU-CLOUD-FACTS:END -->
</div>
</div>
</section>
<!-- ── THE STUDIO: one explained mini video per job ────────── -->
<section class="things hp-things" id="things">
<div class="container">
<div class="hp-head">
<p class="eyebrow">What you can do</p>
<h2>Not a chat app. <em class="rx">The whole studio.</em></h2>
<p class="section-lead">Real AI on your own computer. One installer, no command line, no Docker. Private, free, and yours. In local mode there is no account, no cloud and nobody watching.</p>
</div>
<article class="ft" id="chat">
<div class="ft-copy">
<span class="ft-num"><b>01</b> Chat</span>
<h3 class="ft-h">Uncensored AI chat, like ChatGPT <em>but yours.</em></h3>
<p>Ask anything and you get an answer instead of a lecture. Show it a photo, drop in a PDF, and it remembers what matters across chats. Ask for a picture and it draws one right in the conversation. The app finds and installs the right free model for your PC, so all you do is type. No PC for it? There is <a href="/uncensored-ai-chat-online/">uncensored AI chat online</a> too.</p>
<ol class="mv-ch" data-mv-for="mv-chat">
<li><button type="button" data-i="0">Ask anything</button></li>
<li><button type="button" data-i="1">A filtered model refuses</button></li>
<li><button type="button" data-i="2">LU answers</button></li>
<li><button type="button" data-i="3">Ask for a picture</button></li>
</ol>
<div class="thing-tags"><span>Qwen 3.8</span><span>GPT-OSS</span><span>GLM-4.7</span><span>DeepSeek R1</span></div>
</div>
<figure class="mv" id="mv-chat" data-mv="chat">
<div class="mv-top"><span class="mv-live"><i></i>Live demo</span><span class="mv-name">Chat · no refusal</span></div>
<div class="mv-box" data-nosnippet style="--mv-poster:url(/assets/home/nun.webp)"><div class="mv-clip"><div class="mv-stage" aria-hidden="true" translate="no"></div></div></div>
<div class="mv-ctl"><button type="button" class="mv-pp" aria-label="Pause demo"></button><div class="mv-track"><i class="mv-fill"></i></div><span class="mv-tc">0:00</span></div>
<figcaption class="sr-only">Demo: a user asks for a disturbing horror scene. A typical filtered model refuses, Locally Uncensored writes it, then draws the character right in the chat.</figcaption>
</figure>
</article>
<article class="ft ft--flip" id="create">
<div class="ft-copy">
<span class="ft-num"><b>02</b> Create</span>
<h3 class="ft-h">Images and video <em>from a sentence.</em></h3>
<p>Type what you imagine and get a picture. Change it with words, cut it out, or turn it into a clip. Twelve modes share one tab, and everything you make lands in your own gallery. The app only offers what your computer can handle and installs each tool with one click.</p>
<div class="mv-modes" data-mv-for="mv-create">
<button type="button" data-i="0">Image</button><button type="button" data-i="1">Edit</button><button type="button" data-i="2">Cutout</button><button type="button" data-i="3">Enhance</button><button type="button" data-i="4">Erase</button><button type="button" data-i="5">Video</button><button type="button" data-i="6">Animate</button><button type="button" data-i="7">Character</button><button type="button" data-i="8">Lipsync</button><button type="button" data-i="9">Music</button><button type="button" data-i="10">Extend</button><button type="button" data-i="11">Motion</button>
</div>
<p class="ft-note">All twelve run on LU Cloud. Enhance and Erase need the cloud; the other ten also run on your own graphics card.</p>
<div class="thing-tags"><span>FLUX 2</span><span>Juggernaut XL</span><span>LTX 2.3</span><span>Wan 2.1</span></div>
</div>
<figure class="mv" id="mv-create" data-mv="create">
<div class="mv-top"><span class="mv-live"><i></i>Live demo</span><span class="mv-name">Create · 12 modes</span></div>
<div class="mv-box" data-nosnippet style="--mv-poster:url(/assets/home/neon.webp)"><div class="mv-clip"><div class="mv-stage" aria-hidden="true" translate="no"></div></div></div>
<div class="mv-ctl"><button type="button" class="mv-pp" aria-label="Pause demo"></button><div class="mv-track"><i class="mv-fill"></i></div><span class="mv-tc">0:00</span></div>
<figcaption class="sr-only">Demo: the Create tab runs through all twelve modes, from a new image and an edit to a cutout, a video clip, a talking character and a music track.</figcaption>
</figure>
</article>
<article class="ft" id="code">
<div class="ft-copy">
<span class="ft-num"><b>03</b> Code</span>
<h3 class="ft-h">It finds the bug <em>and shows you the fix.</em></h3>
<p>Point the Coding Agent at your project and describe the problem in plain words. It reads the files, says what it found, and shows every change as a diff before anything touches your code. Here it catches a temperature formula that is off by 0.15. It even runs on small models most PCs can handle.</p>
<ol class="mv-ch" data-mv-for="mv-code">
<li><button type="button" data-i="0">Describe the problem</button></li>
<li><button type="button" data-i="1">It reads your files</button></li>
<li><button type="button" data-i="2">You see the diff first</button></li>
<li><button type="button" data-i="3">Apply, then it tests</button></li>
</ol>
<div class="thing-tags"><span>Coding Agent</span><span>Claude Code</span><span>14 tools</span><span>MCP</span></div>
</div>
<figure class="mv" id="mv-code" data-mv="code">
<div class="mv-top"><span class="mv-live"><i></i>Live demo</span><span class="mv-name">Coding Agent · diff first</span></div>
<div class="mv-box" data-nosnippet><div class="mv-clip"><div class="mv-stage" aria-hidden="true" translate="no"></div></div></div>
<div class="mv-ctl"><button type="button" class="mv-pp" aria-label="Pause demo"></button><div class="mv-track"><i class="mv-fill"></i></div><span class="mv-tc">0:00</span></div>
<figcaption class="sr-only">Demo: the Coding Agent reads the source file, finds a conversion that adds 32.15 instead of 32, shows the one line diff, applies it and runs the tests.</figcaption>
</figure>
</article>
<article class="ft ft--flip" id="agent">
<div class="ft-copy">
<span class="ft-num"><b>04</b> Agent Mode</span>
<h3 class="ft-h">It stops talking <em>and starts doing.</em></h3>
<p>Give Agent Mode a goal and it picks its own tools. It searches the web, writes the files and runs each step while you watch. Ask for a website and you get one, built in a sandbox on your own machine.</p>
<ol class="mv-ch" data-mv-for="mv-agent">
<li><button type="button" data-i="0">Give it a goal</button></li>
<li><button type="button" data-i="1">It picks its tools</button></li>
<li><button type="button" data-i="2">It writes the files</button></li>
<li><button type="button" data-i="3">It runs the result</button></li>
</ol>
<div class="thing-tags"><span>web_search</span><span>file_write</span><span>shell_execute</span><span>delegate_task</span></div>
</div>
<figure class="mv" id="mv-agent" data-mv="agent">
<div class="mv-top"><span class="mv-live"><i></i>Live demo</span><span class="mv-name">Agent Mode · a website from one sentence</span></div>
<div class="mv-box" data-nosnippet><div class="mv-clip"><div class="mv-stage" aria-hidden="true" translate="no"></div></div></div>
<div class="mv-ctl"><button type="button" class="mv-pp" aria-label="Pause demo"></button><div class="mv-track"><i class="mv-fill"></i></div><span class="mv-tc">0:00</span></div>
<figcaption class="sr-only">Demo: Agent Mode gets one sentence about a bakery website, searches the web, writes index.html and style.css, and starts a local server while the page builds up next to it.</figcaption>
</figure>
</article>
<article class="ft" id="phone">
<div class="ft-copy">
<span class="ft-num"><b>05</b> Remote</span>
<h3 class="ft-h">Use it <em>from your phone.</em></h3>
<p>Your PC does the heavy lifting and your phone is the remote. Scan a QR code, type a 6-digit code, and chat, coding and tools are in your pocket, at home or on the road. Remote access stays off until you switch it on.</p>
<ol class="mv-ch" data-mv-for="mv-phone">
<li><button type="button" data-i="0">Scan the QR code</button></li>
<li><button type="button" data-i="1">Type the 6-digit code</button></li>
<li><button type="button" data-i="2">Your PC does the work</button></li>
</ol>
<div class="thing-tags"><span>LAN</span><span>tunnel</span><span>QR</span><span>mobile</span></div>
</div>
<figure class="mv" id="mv-phone" data-mv="phone">
<div class="mv-top"><span class="mv-live"><i></i>Live demo</span><span class="mv-name">Remote access · phone</span></div>
<div class="mv-box" data-nosnippet style="--mv-poster:url(/assets/home/alien.webp)"><div class="mv-clip"><div class="mv-stage" aria-hidden="true" translate="no"></div></div></div>
<div class="mv-ctl"><button type="button" class="mv-pp" aria-label="Pause demo"></button><div class="mv-track"><i class="mv-fill"></i></div><span class="mv-tc">0:00</span></div>
<figcaption class="sr-only">Demo: a phone scans the QR code on the PC, enters the 6-digit passcode and asks for a picture; the PC renders it and the phone shows the result.</figcaption>
</figure>
</article>
<article class="ft ft--flip" id="voice">
<div class="ft-copy">
<span class="ft-num"><b>06</b> Voice</span>
<h3 class="ft-h">Talk to it. <em>It answers out loud.</em></h3>
<p>Click Start voice and just talk. LU hears when you are done and answers in one of fifteen voices, and you can speak over the answer at any time. Speech recognition and the voice run on your own computer, and the session is saved as a normal chat. Voice is in beta and speaks English.</p>
<ol class="mv-ch" data-mv-for="mv-voice">
<li><button type="button" data-i="0">Click Start voice</button></li>
<li><button type="button" data-i="1">Just talk</button></li>
<li><button type="button" data-i="2">It answers out loud</button></li>
<li><button type="button" data-i="3">Speak over it</button></li>
</ol>
<div class="thing-tags"><span>Beta</span><span>English</span><span>15 voices</span><span>runs locally</span></div>
</div>
<figure class="mv" id="mv-voice" data-mv="voice">
<div class="mv-top"><span class="mv-live"><i></i>Live demo</span><span class="mv-name">Voice · beta</span></div>
<div class="mv-box" data-nosnippet><div class="mv-clip"><div class="mv-stage" aria-hidden="true" translate="no"></div></div></div>
<div class="mv-ctl"><button type="button" class="mv-pp" aria-label="Pause demo"></button><div class="mv-track"><i class="mv-fill"></i></div><span class="mv-tc">0:00</span></div>
<figcaption class="sr-only">Demo: a click on Start voice, a spoken question about dinner, a spoken answer with the transcript next to it, and the user speaking over the answer to change it.</figcaption>
</figure>
</article>
<article class="ft" id="computer-control">
<div class="ft-copy">
<span class="ft-num"><b>07</b> Computer control</span>
<h3 class="ft-h">It uses your mouse and keyboard. <em>Only where you allow it.</em></h3>
<p>Switch computer control on and a model can look at your screen and click and type in the apps you approve. It asks once in every chat, once for every app and, unless you choose otherwise, before every action. The LU window shrinks to a small panel with Pause and Stop, and you see every move. Computer control is in beta and off until you switch it on. On a Mac it is not fully tested yet.</p>
<ol class="mv-ch" data-mv-for="mv-computer">
<li><button type="button" data-i="0">Ask in plain words</button></li>
<li><button type="button" data-i="1">You allow it first</button></li>
<li><button type="button" data-i="2">It asks for each app</button></li>
<li><button type="button" data-i="3">You confirm, it acts</button></li>
</ol>
<div class="thing-tags"><span>Beta</span><span>off by default</span><span>asks first</span><span>Pause and Stop</span></div>
</div>
<figure class="mv" id="mv-computer" data-mv="computer">
<div class="mv-top"><span class="mv-live"><i></i>Live demo</span><span class="mv-name">Computer control · beta</span></div>
<div class="mv-box" data-nosnippet><div class="mv-clip"><div class="mv-stage" aria-hidden="true" translate="no"></div></div></div>
<div class="mv-ctl"><button type="button" class="mv-pp" aria-label="Pause demo"></button><div class="mv-track"><i class="mv-fill"></i></div><span class="mv-tc">0:00</span></div>
<figcaption class="sr-only">Demo: a user asks LU to write a shopping list into an open Notepad window. LU asks whether it may use computer control in this chat, shrinks to a small panel, asks before it touches Notepad, asks again before it types, then writes the list and comes back.</figcaption>
</figure>
</article>
</div>
</section>
<!-- ── WHAT'S NEW (names no version, links to the latest release) ────────────────────── -->
<section class="models" id="whats-new">
<div class="container">
<div class="hp-head">
<p class="eyebrow">What's new</p>
<h2>Every release says <em class="rx">what changed</em>.</h2>
<p class="section-lead">Each release comes with notes that say what was added, what was fixed and on which platform. The app shows them after an update, and the same notes are on GitHub next to the installers. Updates are free, as always: the app fetches a new release itself, and installing it takes one click.</p>
</div>
<p class="blog-all"><a href="https://github.com/PurpleDoubleD/locally-uncensored/releases/latest">Notes for the latest release &rarr;</a> <a href="https://github.com/PurpleDoubleD/locally-uncensored/blob/master/CHANGELOG.md">Read the full changelog &rarr;</a> <a href="/blog/">Release posts on the blog &rarr;</a></p>
</div>
</section>
<!-- ── HOW IT WORKS ──────────────────────────────────────── -->
<section class="how hp-how" id="how">
<div class="container">
<div class="hp-head">
<p class="eyebrow">Setup</p>
<h2>Running in under <em class="rx">five minutes.</em></h2>
<p class="section-lead">No tech skills needed. There is no command line and there are no setup files. Install it like any normal app and you're chatting in minutes.</p>
</div>
<div class="how-split">
<div class="steps" data-mv-for="mv-setup">
<article class="step" data-i="0" tabindex="0">
<span class="step-num">01 &middot; Install</span>
<h3>Download &amp; install.</h3>
<p>One file. Click it and install like any app. It keeps itself up to date. Free, forever.</p>
</article>
<article class="step" data-i="1" tabindex="0">
<span class="step-num">02 &middot; Detect</span>
<h3>Wizard finds everything.</h3>
<p>On first start it checks what you've already got and sets it up for you. Nothing installed yet? It grabs the free AI engine for you in one click.</p>
</article>
<article class="step" data-i="2" tabindex="0">
<span class="step-num">03 &middot; Run</span>
<h3>Chat. Code. Create.</h3>
<p>Pick an AI and start typing. Flip to coding, or to images and video, whenever you like. Want to add a cloud account later? One setting.</p>
</article>
</div>
<figure class="mv" id="mv-setup" data-mv="setup">
<div class="mv-top"><span class="mv-live"><i></i>Live demo</span><span class="mv-name">First start</span></div>
<div class="mv-box" data-nosnippet><div class="mv-clip"><div class="mv-stage" aria-hidden="true" translate="no"></div></div></div>
<div class="mv-ctl"><button type="button" class="mv-pp" aria-label="Pause demo"></button><div class="mv-track"><i class="mv-fill"></i></div><span class="mv-tc">0:00</span></div>
<figcaption class="sr-only">Demo: the installer runs, the first start wizard checks twelve local AI engines, and the app opens with a model that fits the PC.</figcaption>
</figure>
</div>
<p class="guide-link"><a href="https://locallyuncensored.com/guide/">The full handbook: install, first start, chat, agent, Create, Cloud &rarr;</a></p>
</div>
</section>
<!-- ── MODELS ────────────────────────────────────────────── -->
<section class="models hp-models" id="models">
<div class="container">
<div class="hp-head">
<p class="eyebrow">Models</p>
<h2>The best <em class="rx">free AI</em>, for every job.</h2>
<p class="section-lead">Whatever you want to do, whether that is chat, art or video, there's a free AI for it. The app installs them for you and only suggests the ones your computer can run. The big names are all here: <strong>Qwen</strong>, <strong>GPT-OSS</strong>, <strong>Llama 4</strong>, <strong>Gemma 4</strong>, <strong>DeepSeek</strong> and more.</p>
</div>
<div class="model-grid">
<article class="model-card">
<span class="model-type">CHAT &middot; VISION</span>
<h3>Gemma 4</h3>
<p>Google's free chat AI. It can read your pictures too, and the small version runs on a basic PC.</p>
<span class="model-meta">E4B &middot; 27B</span>
</article>
<article class="model-card">
<span class="model-type">CHAT &middot; REASONING</span>
<h3>Qwen 3.8 &middot; GPT-OSS &middot; GLM-4.7</h3>
<p>The smart ones, best for thinking through tough questions and writing code. Uncensored versions are included that just answer, with no lectures.</p>
<span class="model-meta">Uncensored versions &middot; gaming PC</span>
</article>
<article class="model-card">
<span class="model-type">IMAGE</span>
<h3>Juggernaut XL</h3>
<p>A crowd favourite for photo-real images. Make a picture from words, or remix one you already have.</p>
<span class="model-meta">Photo-real &middot; gaming PC</span>
</article>
<article class="model-card">
<span class="model-type">IMAGE &middot; UNCENSORED</span>
<h3>Z-Image Turbo</h3>
<p>Truly uncensored, no filters. Loads slowly below 16 GB.</p>
<span class="model-meta">Needs a strong graphics card</span>
</article>
<article class="model-card">
<span class="model-type">VIDEO &middot; TEXT-TO-VIDEO</span>
<h3>LTX 2.3</h3>
<p>Turn a sentence into a video clip. Quick, and works on regular hardware.</p>
<span class="model-meta">Text to video &middot; gaming PC</span>
</article>
<article class="model-card">
<span class="model-type">VIDEO &middot; IMAGE-TO-VIDEO</span>
<h3>FramePack F1</h3>
<p>Got a picture? Turn it into a moving video. Runs on most laptops with a graphics card.</p>
<span class="model-meta">Photo to video &middot; most laptops</span>
</article>
</div>
</div>
</section>
<!-- ── MANIFESTO ──────────────────────────────────────────── -->
<section class="manifesto hp-manifesto">
<div class="narrow">
<p class="eyebrow">Why</p>
<h2>Made for people who want <em class="rx">their</em> AI on <em class="rx">their</em> machine.</h2>
<p>In local mode your chats never leave your computer. No subscription, no limits, and no company reading along. Most local AI apps stop at text chat; this one gives you <em>chat, code, images, and video</em> in one place, and your phone can run it from anywhere.</p>
<p>It's <strong>free</strong> and <strong>open source</strong>, so anyone can check the code. Local wins for privacy and unlimited use; when you want a cloud model for the really hard stuff, you can plug your own in. Tested with <strong>2,200+ automated checks</strong> so it just works.</p>
</div>
</section>
<!-- ── BLOG ──────────────────────────────────────────────── -->
<section class="blog hp-blog" id="blog">
<div class="container">
<div class="hp-head">
<p class="eyebrow">Writing</p>
<h2>Guides, comparisons, release notes.</h2>
<p class="section-lead">Plain-English guides to running AI on your own PC: what to install, what runs on your machine, and how it compares to the rest.</p>
</div>
<div class="post-list">
<a class="post" href="https://locallyuncensored.com/blog/how-to-run-ai-locally.html">
<span class="post-cat">Beginner</span>
<h3>How to Run AI Locally: 5-Minute Beginner Guide</h3>
<p>No command line, no Docker. One installer, one-click models, honest hardware requirements.</p>
</a>
<a class="post" href="https://locallyuncensored.com/qwen-image-2-1/">
<span class="post-cat">Guide</span>
<h3>Qwen-Image 2.1 in Locally Uncensored</h3>
<p>One local model that draws and edits. The three files, the ComfyUI version it needs and the license.</p>
</a>
<a class="post" href="https://locallyuncensored.com/blog/abliterated-models-guide.html">
<span class="post-cat">Guide</span>
<h3>Abliterated Models Guide</h3>
<p>Qwen 3.6, Gemma 4 Heretic, Llama 3.1, Hermes 3. What abliteration is, where to download.</p>
</a>
<a class="post" href="https://locallyuncensored.com/blog/best-uncensored-ai-models-2026.html">
<span class="post-cat">Guide</span>
<h3>Best Uncensored AI Models in 2026</h3>
<p>Which open models answer without refusing, sorted by what your card can hold.</p>
</a>
<a class="post" href="https://locallyuncensored.com/blog/local-ai-on-your-phone.html">
<span class="post-cat">Guide</span>
<h3>Local AI on Your Phone</h3>
<p>Your PC runs the model, your phone is the remote. Private, free, no APK needed.</p>
</a>
<a class="post" href="https://locallyuncensored.com/blog/image-to-image-local-ai.html">
<span class="post-cat">Guide</span>
<h3>Image-to-Image with Local AI</h3>
<p>Upload a photo and turn it into something new. Powered by FLUX, Z-Image, SDXL.</p>
</a>
<a class="post" href="https://locallyuncensored.com/blog/run-uncensored-ai-locally.html">
<span class="post-cat">Guide</span>
<h3>How to Run Uncensored AI Locally</h3>
<p>Setup guide. Models, hardware, and why local beats cloud.</p>
</a>
<a class="post" href="https://locallyuncensored.com/blog/uncensored-ai-video-locally.html">
<span class="post-cat">Guide</span>
<h3>Uncensored AI Video on Your Own GPU</h3>
<p>Text to video and image to video without a content filter in between.</p>
</a>
<a class="post" href="https://locallyuncensored.com/qwen-3-8-uncensored/">
<span class="post-cat">Guide</span>
<h3>Qwen 3.8 Uncensored</h3>
<p>Abliterated and heretic builds of Qwen 3.8 27B, and which GGUF file fits your VRAM.</p>
</a>
<a class="post" href="https://locallyuncensored.com/blog/uncensored-ai-chat.html">
<span class="post-cat">Guide</span>
<h3>Uncensored AI Chat, Local and Private</h3>
<p>A chat that runs on your machine, with no account and no log on a server.</p>
</a>
<a class="post" href="https://locallyuncensored.com/blog/comfyui-beginners-guide.html">
<span class="post-cat">Guide</span>
<h3>ComfyUI for Beginners</h3>
<p>How LU handles ComfyUI setup, models, and workflows automatically.</p>
</a>
<a class="post" href="https://locallyuncensored.com/blog/lm-studio-alternatives.html">
<span class="post-cat">Comparison</span>
<h3>7 Best LM Studio Alternatives in 2026</h3>
<p>Locally Uncensored, Jan, Ollama, Open WebUI, GPT4All, Msty and KoboldCpp, compared honestly.</p>
</a>
<a class="post" href="https://locallyuncensored.com/blog/gemma-4-local-guide.html">
<span class="post-cat">Guide</span>
<h3>Google Gemma 4: Run It Locally</h3>
<p>All sizes from E4B to 27B. Native tools, vision, uncensored variants.</p>
</a>
<a class="post" href="https://locallyuncensored.com/blog/easiest-local-ai-image-generator.html">
<span class="post-cat">Beginner</span>
<h3>The Easiest Local AI Image Generator</h3>
<p>FLUX and SDXL quality on your GPU without touching a ComfyUI node graph.</p>
</a>
<a class="post" href="https://locallyuncensored.com/blog/locally-uncensored-vs-lm-studio.html">
<span class="post-cat">Versus</span>
<h3>LU vs LM Studio</h3>
<p>Open source all-in-one vs polished closed-source chat client.</p>
</a>
</div>
<p class="blog-all"><a href="https://locallyuncensored.com/blog/">View all posts &rarr;</a></p>
</div>
</section>
<!-- ── FAQ ───────────────────────────────────────────────── -->
<section class="faq hp-faq" id="faq">
<div class="container narrow">
<p class="eyebrow">Questions</p>
<h2>Common questions.</h2>
<div class="faq-list">
<details class="faq-item" open>
<summary><h3>What is Locally Uncensored?</h3></summary>
<p>A free, open-source local AI studio for your desktop. Install it like any normal app and you're chatting, generating images, and making videos in minutes, with no command line and no Docker. Under the hood: chat with 20+ provider presets, a coding agent, image generation via ComfyUI (FLUX 2, Juggernaut XL, Z-Image, SDXL), and video generation (Wan 2.1, HunyuanVideo, LTX 2.3, FramePack F1) in one interface. AGPL-3.0 licensed.</p>
</details>
<details class="faq-item">
<summary><h3>Does it support Qwen 3.8, GPT-OSS and GLM-4.7?</h3></summary>
<p>Yes. <strong>Qwen 3.8 27B</strong> is in the Model Manager since v2.6.6, and vision works: the image tower comes as a separate mmproj file that the download writes next to the model. Official, abliterated and uncensored builds, plus a 9B distill for small cards. <strong>GPT-OSS-120B</strong> and <strong>GPT-OSS-20B</strong> run via Ollama. <strong>GLM-4.7 Flash</strong> supported through Ollama. Also ready: <strong>DeepSeek R1</strong>, <strong>Llama 4</strong>, <strong>Gemma 4</strong>, <strong>Mistral Small 3</strong>, <strong>Phi 4</strong>.</p>
</details>
<details class="faq-item">
<summary><h3>Can I use it as a ChatGPT or Claude alternative?</h3></summary>
<p>Yes. Locally Uncensored works as a ChatGPT and Claude alternative that runs on your own hardware. Use Qwen 3.8, GPT-OSS, GLM-4.7, DeepSeek R1, Llama 4 or Gemma 4 instead, or add cloud providers (OpenAI, Anthropic, OpenRouter, Groq) alongside the local stack.</p>
</details>
<details class="faq-item">
<summary><h3>Is it really free and offline?</h3></summary>
<p>Yes. After setup and model download, no internet is needed for the local providers. Local use needs no account and has no usage limit. The app has no analytics kit and sends no crash reports. On its own it checks GitHub for updates and counts presses of the Cloud switch. Cloud providers are optional, and the core runs on your own hardware.</p>
</details>
<details class="faq-item">
<summary><h3>How is this different from Open WebUI or LM Studio?</h3></summary>
<p>Those tools handle text chat. Locally Uncensored adds a coding agent with fourteen MCP tools, image generation, video creation, A/B model comparison, local benchmarking, granular permissions, file upload with vision, and thinking mode, all in one app.</p>
</details>
<details class="faq-item">
<summary><h3>What hardware do I need?</h3></summary>
<p>Text chat: 8&nbsp;GB RAM. Image generation: an NVIDIA card, or an AMD card with ROCm on Linux, from 6&nbsp;GB of VRAM with the starter checkpoint, 8 to 10&nbsp;GB for FLUX. Video generation: from 6&nbsp;GB for AnimateDiff and image to video with FramePack F1, from 8&nbsp;GB for the Wan 2.1 starter, 12&nbsp;GB and up for Wan 2.2 and HunyuanVideo. The app auto-detects hardware and recommends models. Windows 10/11 and Linux supported, and Apple Silicon Macs since version 3.0.6.</p>
</details>
<details class="faq-item">
<summary><h3>What does "uncensored" mean?</h3></summary>
<p>Abliterated models with artificial restrictions removed. The AI responds honestly without refusing or adding disclaimers. Combined with local execution, your conversations stay private.</p>
</details>
<details class="faq-item">
<summary><h3>Does remote access leak data?</h3></summary>
<p>Remote access is off until you start it, sits behind a six-digit passcode, and shows when a device is connected. On your own network the chat goes straight from phone to PC. Over the internet Cloudflare relays it: the traffic is encrypted to and from Cloudflare, not end to end, so Cloudflare could read it.</p>
</details>
<details class="faq-item">
<summary><h3>Can I talk to it?</h3></summary>
<p>Yes. Voice is a spoken conversation with the chat model you picked. Speech recognition and the voice run on your own machine, in English, with fifteen voices. Voice is in beta: our check in the real app ran on Windows, and your computer may still show rough edges.</p>
</details>
<details class="faq-item">
<summary><h3>Can it control my computer?</h3></summary>
<p>Only if you switch it on. With computer control a model uses the mouse and keyboard in apps you approve. It asks before it touches an app and, by default, before every action, and it never touches terminals, password managers or password fields. You can pause or stop it at any time. Computer control is in beta: it is not proven in every app yet, so watch what it does and stop it when it goes wrong.</p>
</details>
<details class="faq-item">
<summary><h3>Can I use this on macOS?</h3></summary>
<p>Yes, since version 3.0.6, in beta. The Mac build is for Apple Silicon and needs macOS 14 or newer. Intel Macs are not supported. The .dmg is signed and notarized. Chat runs on local models through the bundled server with Metal, and Create makes images and video on the Mac itself, without ComfyUI. We tested it on an M5 Pro with 48 GB. Tell us how it runs on yours. Voice has been tried by hand in the app on a Mac and works. Computer control is in the Mac app but not fully tested yet on macOS. Remote Access and the automatic update are not confirmed on macOS. On any Mac you can also use the hosted studio in a browser at lu-labs.ai.</p>
</details>
</div>
</div>
</section>
<!-- ── GET IT ────────────────────────────────────────────── -->
<section class="get hp-get" id="get">
<div class="narrow">
<svg class="get-mark" viewBox="176 176 672 672" aria-hidden="true"><rect class="get-mark-r" x="193" y="193" width="638" height="638" rx="138" ry="138" fill="none" stroke="currentColor" stroke-width="30" pathLength="1"/><path class="get-mark-p" fill="currentColor" d="M302,342 H377 V612 H433.6 A162,162 0 0,0 500.5,680 H302 Z"/><path class="get-mark-p" fill="currentColor" d="M434,342 H508 V545 A69,69 0 0,0 646,545 V342 H722 V537.5 A144,144 0 0,1 434,537.5 Z"/></svg>
<h2 class="get-h">The whole studio. <em class="rx">Free.</em></h2>
<p class="get-sub">Chat, create and code. On your PC or in the cloud.</p>
<div class="hx-phone">
<a class="dl-btn" href="https://lu-labs.ai/app?demo=1&amp;src=luc-end-phone"><svg width="14" height="14" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2.4" stroke-linecap="round" stroke-linejoin="round" aria-hidden="true"><path d="M5 12h14M13 6l6 6-6 6"/></svg>Try LU Cloud in this browser</a>
<p class="hx-phone-sub">The demo needs no account. Credit packs from 15&nbsp;EUR, plans from 19&nbsp;EUR a month.</p>
<button type="button" class="hx-phone-share" data-share-desktop><svg width="14" height="14" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2.2" stroke-linecap="round" stroke-linejoin="round" aria-hidden="true"><path d="M4 12v7a2 2 0 0 0 2 2h12a2 2 0 0 0 2-2v-7"/><path d="M16 6l-4-4-4 4"/><path d="M12 2v13"/></svg><span>Desktop app for Windows, Mac and Linux: send the link to yourself</span></button>
<p class="hx-phone-done" data-share-status role="status" aria-live="polite"></p>
</div>
<div class="dl-pick">
<div class="dl-card dl-card--xl">
<span class="dl-win" aria-hidden="true"><svg data-dl-for="win" viewBox="0 0 24 24"><path d="M3 5.5 10.5 4.4v7.1H3zM11.5 4.3 21 3v8.5h-9.5zM3 12.5h7.5v7.1L3 18.5zM11.5 12.5H21V21l-9.5-1.3z"/></svg><svg data-dl-for="mac" viewBox="0 0 24 24"><path d="M12.152 6.896c-.948 0-2.415-1.078-3.96-1.04-2.04.027-3.91 1.183-4.961 3.014-2.117 3.675-.546 9.103 1.519 12.09 1.013 1.454 2.208 3.09 3.792 3.039 1.52-.065 2.09-.987 3.935-.987 1.831 0 2.35.987 3.96.948 1.637-.026 2.676-1.48 3.676-2.948 1.156-1.688 1.636-3.325 1.662-3.415-.039-.013-3.182-1.221-3.22-4.857-.026-3.04 2.48-4.494 2.597-4.559-1.429-2.09-3.623-2.324-4.39-2.376-2-.156-3.675 1.09-4.61 1.09zM15.53 3.83c.843-1.012 1.4-2.427 1.245-3.83-1.207.052-2.662.805-3.532 1.818-.78.896-1.454 2.338-1.273 3.714 1.338.104 2.715-.688 3.559-1.701"/></svg><svg data-dl-for="linux" viewBox="3 1.5 18 21"><path fill-rule="evenodd" d="M12 2c-2.2 0-3.6 1.8-3.6 4.2 0 1.5.2 2.4-.6 3.8C6.6 12.2 5 14.6 5 17.4c0 1.2.4 2.2 1 3-.6.3-1 .8-1 1.1 0 .4.6.5 1.6.5h10.8c1 0 1.6-.1 1.6-.5 0-.3-.4-.8-1-1.1.6-.8 1-1.8 1-3 0-2.8-1.6-5.2-2.8-7.4-.8-1.4-.6-2.3-.6-3.8C15.6 3.8 14.2 2 12 2zm0 8.5c1.8 0 3.6 3 3.6 6 0 2-1.4 3.2-3.6 3.2s-3.6-1.2-3.6-3.2c0-3 1.8-6 3.6-6zM10.6 5.5a.7.7 0 1 0 0 1.4.7.7 0 0 0 0-1.4zm2.8 0a.7.7 0 1 0 0 1.4.7.7 0 0 0 0-1.4z"/></svg></span>
<div class="dl-info">
<h3>Locally Uncensored <em>&middot; 3.0.7</em></h3>
<p>v3.0.7 &middot; <span data-dl-for="win">WINDOWS</span> <span data-dl-for="mac">MACOS (BETA) &middot; APPLE SILICON</span> <span data-dl-for="linux">LINUX</span> &middot; <span class="dl-nb">170,000+ DOWNLOADS</span> &middot; 100% open source</p>
</div>
<a class="dl-btn" data-dl-for="win" href="https://github.com/PurpleDoubleD/locally-uncensored/releases/latest/download/Locally.Uncensored_3.0.7_x64-setup.exe" download><svg width="14" height="14" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2.4" stroke-linecap="round" stroke-linejoin="round" aria-hidden="true"><path d="M21 15v4a2 2 0 0 1-2 2H5a2 2 0 0 1-2-2v-4"/><polyline points="7 10 12 15 17 10"/><line x1="12" y1="15" x2="12" y2="3"/></svg>Download for Windows (.exe)</a>
<a class="dl-btn" data-dl-for="mac" href="https://github.com/PurpleDoubleD/locally-uncensored/releases/latest/download/LU_3.0.7_aarch64.dmg" download><svg width="14" height="14" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2.4" stroke-linecap="round" stroke-linejoin="round" aria-hidden="true"><path d="M21 15v4a2 2 0 0 1-2 2H5a2 2 0 0 1-2-2v-4"/><polyline points="7 10 12 15 17 10"/><line x1="12" y1="15" x2="12" y2="3"/></svg>Download for Mac (.dmg)</a>
<a class="dl-btn" data-dl-for="linux" href="https://github.com/PurpleDoubleD/locally-uncensored/releases/latest/download/Locally.Uncensored_3.0.7_amd64.AppImage" download><svg width="14" height="14" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2.4" stroke-linecap="round" stroke-linejoin="round" aria-hidden="true"><path d="M21 15v4a2 2 0 0 1-2 2H5a2 2 0 0 1-2-2v-4"/><polyline points="7 10 12 15 17 10"/><line x1="12" y1="15" x2="12" y2="3"/></svg>Download for Linux (AppImage)</a>
</div>
<p class="dl-android"><a href="https://github.com/PurpleDoubleD/locally-uncensored/releases/latest/download/LU-3.0.7-Android.apk" download>Android phone? Download the APK</a> &middot; Android 7.0 or newer, 64-bit, 27 MB &middot; <a href="/android/">how to install</a></p>
<div class="dl-os" role="group" aria-label="Choose your system">
<a data-dl-os="win" href="https://github.com/PurpleDoubleD/locally-uncensored/releases/latest/download/Locally.Uncensored_3.0.7_x64-setup.exe" download aria-current="true">Windows</a>
<a data-dl-os="mac" href="https://github.com/PurpleDoubleD/locally-uncensored/releases/latest/download/LU_3.0.7_aarch64.dmg" download>macOS</a>
<a data-dl-os="linux" href="https://github.com/PurpleDoubleD/locally-uncensored/releases/latest/download/Locally.Uncensored_3.0.7_amd64.AppImage" download>Linux</a>
</div>
<p class="dl-note">
<span data-dl-note="win">Windows 10 and 11, x64.</span>
<span data-dl-note="mac">macOS is in beta: Apple Silicon, macOS 14 or newer. Intel Macs are not supported. Details on the <a href="/mac/">Mac page</a>.</span>
<span data-dl-note="linux">Linux x64: the AppImage runs on any distribution, or take the <a href="https://github.com/PurpleDoubleD/locally-uncensored/releases/latest/download/Locally.Uncensored_3.0.7_amd64.deb" download>.deb</a> or the <a href="https://github.com/PurpleDoubleD/locally-uncensored/releases/latest/download/Locally.Uncensored-3.0.7-1.x86_64.rpm" download>.rpm</a>.</span>
</p>
</div>
<div class="dl-card dl-card--xl dl-card--cloud">
<span class="dl-win dl-lu" aria-hidden="true"><svg viewBox="193 193 638 638"><rect x="193" y="193" width="638" height="638" rx="138" ry="138" fill="#8b5cf6"/><path fill="#fff" d="M302,342 H377 V612 H433.6 A162,162 0 0,0 500.5,680 H302 Z"/><path fill="#fff" d="M434,342 H508 V545 A69,69 0 0,0 646,545 V342 H722 V537.5 A144,144 0 0,1 434,537.5 Z"/></svg></span>
<div class="dl-info">
<h3>LU Cloud <em>&middot; in your browser</em></h3>
<p>NO GPU NEEDED &middot; MAC &middot; PHONE &middot; PACKS FROM 15 EUR</p>
</div>
<a class="dl-btn" href="https://lu-labs.ai/pricing?tab=credits&amp;src=luc-home-end">
<svg width="14" height="14" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2.4" stroke-linecap="round" stroke-linejoin="round" aria-hidden="true"><path d="M5 12h14M13 6l6 6-6 6"/></svg>
See plans and packs
</a>
</div>
<div class="dl-alts">
<a href="https://locallyuncensored.com/discord" target="_blank" rel="noopener">Join Discord</a>
<a href="https://github.com/PurpleDoubleD/locally-uncensored/releases">All releases</a>
<a href="https://github.com/PurpleDoubleD/locally-uncensored">Source on GitHub</a>
<a href="https://github.com/PurpleDoubleD/locally-uncensored/discussions">Discussions</a>
<a href="https://github.com/PurpleDoubleD/locally-uncensored/blob/master/SECURITY.md#antivirus--browser-false-positives">Is it safe? Why Windows may warn you &rarr;</a>
</div>
</div>
</section>
<footer>
<div class="footer-row footer-meta">
<span>&copy; 2026 &middot; Locally Uncensored &middot; AGPL-3.0</span>
</div>
<div class="footer-row footer-links">
<a href="https://locallyuncensored.com/blog/">Blog</a>
<a href="https://locallyuncensored.com/guide/">Guide</a>
<a href="https://locallyuncensored.com/pricing/">Pricing</a>
<a href="https://locallyuncensored.com/cloud/">LU Cloud</a>
<a href="https://locallyuncensored.com/uncensored-ai-chat-online/">Uncensored AI chat online</a>
<a href="https://locallyuncensored.com/uncensored-ai-image-generator-online/">Image generator online</a>
<a href="https://locallyuncensored.com/uncensored-ai-video-generator-online/">Video generator online</a>
<a href="https://locallyuncensored.com/alternatives/">Alternatives</a>
<a href="https://locallyuncensored.com/is-locally-uncensored-safe/">Is it safe?</a>
<a href="https://locallyuncensored.com/remote-access/">Remote access</a>
<a href="https://locallyuncensored.com/discord" target="_blank" rel="noopener">Discord</a>
<a href="https://github.com/PurpleDoubleD/locally-uncensored">GitHub</a>
<a href="https://github.com/PurpleDoubleD/locally-uncensored/discussions">Discussions</a>
<a href="https://github.com/PurpleDoubleD/locally-uncensored/issues">Issues</a>
<a href="https://reddit.com/user/GroundbreakingMall54">Reddit</a>
<a href="https://www.linkedin.com/in/david-drobnitza-locallyuncensored/" rel="me noopener" target="_blank">LinkedIn</a>
<a href="https://locallyuncensored.com/privacy/">Privacy</a>
<a href="https://locallyuncensored.com/terms/">Terms</a>
<a href="https://locallyuncensored.com/impressum/">Impressum</a>
</div>
</footer>
<script>
// Phones: the share button sends the home page to the visitor's own
// computer. Without the share sheet the link is copied, and if that fails
// too, it is shown so it can be copied by hand. No form, nothing stored.
(function () {
var url = 'https://locallyuncensored.com/';
function say(btn, text) {
var out = btn.parentNode.querySelector('[data-share-status]');
if (out) out.textContent = text;
}
document.addEventListener('click', function (event) {
var btn = event.target.closest ? event.target.closest('[data-share-desktop]') : null;
if (!btn) return;
var data = { title: 'Locally Uncensored', text: 'Locally Uncensored, the desktop app for Windows, Mac and Linux', url: url };
if (navigator.share) {
navigator.share(data).catch(function (err) { if (!err || err.name !== 'AbortError') copy(btn); });
} else copy(btn);
});
function copy(btn) {
if (navigator.clipboard && navigator.clipboard.writeText) {
navigator.clipboard.writeText(url).then(function () { say(btn, 'Link copied. Open it on your computer to download.'); }, function () { say(btn, 'Open ' + url + ' on your computer to download.'); });
} else say(btn, 'Open ' + url + ' on your computer to download.');
}
})();
</script>
<script>
// Always start at top on fresh load (strip lingering hash)
if (location.hash) {
history.replaceState(null, '', location.pathname);
window.scrollTo(0, 0);
}
if ('scrollRestoration' in history) history.scrollRestoration = 'manual';
</script>
</body>
</html>