Locally Uncensored vs Open WebUI

Choosing between local AI frontends? Open WebUI and Locally Uncensored both publish their code and both run AI locally. Locally Uncensored uses AGPL-3.0, Open WebUI its own Open WebUI License. They solve different problems.

Open WebUI is one of the most popular Ollama frontends, with more than 150K GitHub stars. It's battle-tested, feature-rich, and has a massive community. Locally Uncensored is newer and takes a different approach: instead of being the best chat UI, it is a desktop app that combines chat, image generation, video creation and a coding agent in one window.

Open WebUI needs an engine underneath, and Locally Uncensored works with one you already run or ships its own. LU Labs compares the two usual candidates in its Ollama vs LM Studio comparison.

This comparison helps you decide which one fits your needs.

Feature Comparison

Where Locally Uncensored Wins

• Image and video generation: Open WebUI can create and edit images once an admin connects an image engine, and its docs list no video generation. Locally Uncensored installs and runs ComfyUI for you on Windows and Linux, with image models such as FLUX, Juggernaut XL and Qwen-Image 2.1 and video models such as Wan, HunyuanVideo, LTX and AnimateDiff. On a Mac, Create runs on the Mac itself.
• No Docker: run the installer and the app is ready. No Docker Compose files, no container management. Open WebUI recommends Docker; it also installs through pip, and its desktop app is in early alpha.
• Answers directly, abliterated models available: pick a model with the refusal behaviour removed from the weights, or a mainstream one, in the same picker. Nothing sits between you and the model on your own machine.
• Simpler setup: a first-run wizard finds an engine you already run, or installs one with a single click, and does the same for ComfyUI. The model manager marks which models fit your hardware.
• Coding agent: it builds a map of the repo, edits only the lines you asked for, shows the diff before it applies anything, and runs your tests. Open WebUI has no coding agent of this kind; its Open Terminal lets a model run commands and code.
• Agent tools and MCP: web search and fetch, file operations, a shell, code execution, screenshots and image and video generation are built in, with no plugins or pipelines needed, and your own MCP servers add more. Open WebUI connects MCP servers natively as well.
• Permissions: every tool call passes a permission gate you control, and a read only run stays read only. In Open WebUI, approving tool calls is an experimental option.

Version 3.0.6 also brings phone access over your LAN or a tunnel, side by side model compare and a local benchmark, a Local API that serves your local models on one OpenAI compatible address, and two beta features: Voice, a spoken conversation run on your own machine, and computer control, which is off until you switch it on.

What Locally Uncensored 3.0.7 adds

Version 3.0.7 adds these points that matter next to Open WebUI:

• ComfyUI workflows in the normal save format: the Workflows view imports a workflow saved with Save or Export in ComfyUI, one file or a whole folder, and Create then shows every node setting of it to change. Node packs install, update and move to the trash from Settings, AI Backends. Windows and Linux only; a Mac runs no ComfyUI workflows. Subgraphs and updating a node pack have run in automated tests only.
• Several local servers at once: KoboldCpp, LM Studio, a llama.cpp server or any other OpenAI compatible backend each get their own row in Settings, AI Backends, and their models sit in the picker under that name. Adding one no longer pushes another out, and a group chat can mix models from several of them. The context window comes from each server itself. So far this has run in automated tests only, not yet in the app with two real servers.
• Duplicate and branch: Duplicate chat copies a conversation with its persona and settings, Branch from here starts a new chat from any message, and Regenerate keeps the earlier answer as a version you can switch back to. These have run in automated tests only.
• Character cards: a card as PNG or JSON (versions 1, 2 and 3) becomes a persona with the card picture. A new chat opens with the card's greeting, alternate greetings switch with arrows, and a card persona exports back as a version 2 JSON file. Checked in the app with Holo and Qwen models, not with Gemma.
• Reasoning effort for local models: the Low, Medium, High and Max button next to Think now works on the LU Engine, on LM Studio 0.4.8 and later, on KoboldCpp (Low to High) and on a llama.cpp server whose chat template reads it. It is one setting for all models, and it has run in automated tests only, not against a real LM Studio, KoboldCpp or llama.cpp server. A separate think budget in tokens works with the LU Engine and with Anthropic.
• Agent safety: text from a page the agent fetches reaches the model marked as page content, not as instructions, and the LU Engine only runs models for the app that started it.

The Settings, Local API page still reaches only the LU Engine, Ollama and LM Studio. Our check of 3.0.7 in the real app ran on one Windows machine.

Where Open WebUI Wins

• Massive community: more than 150K stars means more contributors, faster bug fixes, and better documentation.
• RAG / document chat: upload PDFs, web pages, and documents. Locally Uncensored has Document Chat with a local index too.
• Multi-user support: share one instance with your team or household with role-based access. Locally Uncensored is a single-user desktop app; its phone access is for your own devices.
• Plugin ecosystem: filters, actions, pipes, tools and skills extend Open WebUI's capabilities, and MCP and OpenAPI tool servers plug in too.
• Cloud model support: connect any OpenAI compatible API alongside local Ollama models. Locally Uncensored does this as well.
• Thinking and vision: reasoning shows up in a collapsible block, and vision models take image uploads.

What this page says about Open WebUI comes from its official documentation and GitHub repository, as of October 2026.

More comparisons

• Alternatives and comparisons, every head to head on one page
• Locally Uncensored vs LM Studio, the most asked comparison
• Locally Uncensored vs Ollama, app against engine
• What Locally Uncensored is, what runs on your own machine and what runs on LU Cloud