Repolode
Submit
mrdjohnson

mrdjohnson/llm-x

A browser-based chat UI for local LLM servers such as Ollama, LM Studio and AUTOMATIC1111, delivered as a web app, PWA and Chrome extension.

Open-source alternative to:ChatGPT

View on GitHubWebsite
Stars
296
Forks
38
Open issues
6
Last commit
last month

What is it?

LLM X is a ChatGPT-style interface for running language and image models on your own machine. It connects to local servers such as Ollama, llmman, LM Studio, AUTOMATIC1111 and any OpenAI-compatible server, and it can also use Gemini Nano through Chrome. Chats, images and generations are stored in the browser, and the app makes no external API calls of its own.

The project ships as a hosted web app that works offline as a PWA, as a Chrome extension, and as a source build run with Vite or Docker. Chats can be sent to several models at once, and users can manage multiple conversations, edit or re-run messages, define system prompts through Personas, and import or export chats. Images can be sent to multimodal models, and text-to-image generation is supported through Ollama and AUTOMATIC1111.

The app is built with React, TypeScript, MobX and Tailwind/DaisyUI, and it uses LangChain.js for model interaction. Remote servers must be served over HTTPS, since the web version only connects to secure contexts.

Key features

  • 01Connects to Ollama, llmman, LM Studio, AUTOMATIC1111 and OpenAI-compatible servers.
  • 02Sends one chat to multiple models at the same time.
  • 03Supports image input for multimodal models and text-to-image generation.
  • 04Works offline as a PWA and ships as a Chrome extension.
  • 05Includes Personas for custom system prompts, plus chat import and export.
  • 06Renders code highlighting and LaTeX in responses.

Use cases

  • Chatting with models running on your own laptop or home server without sending data to a cloud service.
  • Comparing answers from several local models side by side in one conversation.
  • Using a local vision model to ask questions about images you upload.
  • Using Chrome's built-in Gemini Nano from a browser extension.

Quick start

  1. 1

    Install Ollama from ollama.com

  2. 2

    Start Ollama with the LLM-X origin allowed

    OLLAMA_ORIGINS=https://mrdjohnson.github.io ollama serve
  3. 3

    Open the LLM-X web app in your browser and start chatting

Things to know

  • The web app will only connect to remote servers that use HTTPS, so remote use needs a secure server.
  • You must install and run a local model server such as Ollama, LM Studio or AUTOMATIC1111 yourself.
  • Gemini Nano support requires enabling experimental Chrome Canary flags.

In collections

Run LLMs locallyAI chatbots & chat clientsChrome & browser extensions

Topics

#llm#ollama#chatbot#local-llm#react#typescript#pwa#chrome-extension#ai-ui#automatic1111

FAQ

What is llm-x?
A browser-based chat UI for local LLM servers such as Ollama, LM Studio and AUTOMATIC1111, delivered as a web app, PWA and Chrome extension. LLM X is a ChatGPT-style interface for running language and image models on your own machine. It connects to local servers such as Ollama, llmman, LM Studio, AUTOMATIC1111 and any OpenAI-compatible server, and it can also use Gemini Nano through Chrome. Chats, images and generations are stored in the browser, and the app makes no external API calls of its own.
Is llm-x free to use? What license does it have?
llm-x is open source under the MIT license. Check the license terms before using it in commercial products.
Is llm-x still maintained?
Updated within 6 months. The last commit to llm-x was on Sep 29, 2026.
What can I use llm-x for?
Chatting with models running on your own laptop or home server without sending data to a cloud service. Comparing answers from several local models side by side in one conversation. Using a local vision model to ask questions about images you upload. Using Chrome's built-in Gemini Nano from a browser extension.
How do I install llm-x?
Install Ollama from ollama.com Start Ollama with the LLM-X origin allowed: OLLAMA_ORIGINS=https://mrdjohnson.github.io ollama serve Open the LLM-X web app in your browser and start chatting
Where can I get llm-x?
The source code and installation instructions are on GitHub: https://github.com/mrdjohnson/llm-x

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects