Llm X
LLMX; Easiest 3rd party Local LLM UI for the web!
Last verified:
What is Llm X?
LLM X is a progressive web application and Chrome extension that provides the easiest third-party Local LLM UI for the web. It serves as a ChatGPT-style interface for users who run large language models locally through Ollama, LM Studio, AUTOMATIC1111, or Google's Gemini Nano. The app works completely offline as a Progressive Web App (PWA) and makes no external API calls, ensuring 100% privacy for all chats and image generations.
Key features include multi-model support allowing simultaneous responses from different models across connected servers, text-to-image generation through AUTOMATIC1111, image-to-text capabilities using Ollama's multi-modal models, code highlighting with Highlight.js, LaTeX support, and a search/command bar (kbar) for quick access to app features. The app supports auto-saved chat history, message editing/deleting/copying, system prompt customization through Personas, theme changing via DaisyUI, chat image modal previews, and import/export functionality for chats.
LLM X is designed for developers, AI enthusiasts, and casual users who want to interact with local LLMs without complex setup. It targets the niche group of folks running Ollama locally who need an offline ChatGPT-style server interface. The app accommodates both casual users wanting quick browser-based interaction and developers needing a unified interface for different AI backends including local setups.
The application supports multiple connections simultaneously, allowing users to connect to Ollama, LM Studio, Open AI servers, and Gemini Nano at the same time. Recent additions include IndexedDB support for storing text, auto-connect on load for visible servers, Chrome extension support with all web app features, and continuous deployment through GitHub Pages.
Llm X pricing
Pricing model: Freemium
Free and open source under MIT License. No paid plans or tiers. The web app at https://mrdjohnson.github.io/llm-x/ is completely free to use. The Chrome extension is also free. All features including multi-model support, image generation, personas, themes, and import/export are included at no cost. Users only need to install the backend services (Ollama, LM Studio, AUTOMATIC1111) which are also free/open source.
Llm X pros
- Completely private with no external API calls
- Works completely offline as a PWA
- Ollama integration for local LLMs
- LM Studio integration supported
- Open AI server integration available
- Gemini Nano on-device model support
- AUTOMATIC1111 integration for image generation
- Text-to-image generation capabilities
- Image-to-text with multi-modal models
- Multi-model support with simultaneous responses
- Code highlighting with Highlight.js
- LaTeX support for mathematical expressions
- Search/command bar (kbar) for quick access
- Auto-saved chat history
- Message editing, copying, and deleting
- System prompt customization via Personas feature
- Theme changing through DaisyUI
- Import and export chat functionality
- Chrome extension version available
- IndexedDB for text storage
- Auto-connect on load for ready servers
- Continuous deployment via GitHub Pages
- Cross-device accessibility through web technologies
- Lightweight with minimal setup requirements
Llm X cons
- Requires HTTPS server for network use (not non-secure contexts)
- Ollama needs custom OLLAMA_ORIGINS environment variable setup
- LM Studio requires terminal command for CORS ($lms server start --cors=true$)
- AUTOMATIC1111 needs special flags for API and CORS
- Chrome extension requires manual OLLAMA_ORIGINS configuration if auto-setup fails
- Code highlighting only handles common languages
- Gemini Nano requires Chrome Canary installation
- Gemini Nano needs BypassPerfRequirement flag enabled
- LangChain.js had issues with stopping incoming streams initially
- Minimum engine requirements specified in package.json
- Blank screen debugging issues noted in contribution guide
- Platform splitting still in progress
- Requires local installation of backend services (Ollama, LM Studio, etc.)
- Web client needs CORS configuration for non-localhost use
- PWA installation required for offline functionality
Frequently asked questions about Llm X
What is LLM X?
LLM X is a ChatGPT-style UI for users who run Ollama locally. It's a progressive web application and Chrome extension that provides an easy-to-use third-party Local LLM UI for the web, supporting both text and image sending/receiving while working completely offline through PWA standards.
Is LLM X private?
Yes, LLM X is 100% private. It does not make any external API calls - you can verify this in your browser's network tab under the Fetch section. All chats and image generations stay completely local and private to your device.
Does LLM X work offline?
Yes, the app works completely offline through PWA (Progressive Web App) standards. You can install it from your browser's download/install button and use it without an internet connection, though you'll need the backend services (Ollama, LM Studio) running locally.
What LLM providers does LLM X support?
LLM X supports Ollama, LM Studio, Open AI servers, Gemini Nano (Google's on-device model), and AUTOMATIC1111 for image generation. You can connect to multiple servers simultaneously and even get responses from different models at the same time.
What is the Chrome extension version?
LLM X has a Chrome extension that includes all features of the web app built directly into the browser. For Ollama with the extension, set OLLAMA_ORIGINS to chrome-extension://iodcdhcpahifeligoegcmcdibdkffclk. The extension ideally works out of the box without special configuration.
How do I use multi-model support?
Multi-model support allows chats to use multiple models from any connected server, and they can respond simultaneously. Simply connect to multiple servers (Ollama, LM Studio, etc.) and select different models for your chats. This feature was added as a recent enhancement.
What Personas feature does?
The Personas feature allows customization of system prompts. You can create different personas with custom system prompts to guide how the LLM responds, giving you control over the AI's personality and response style for different use cases.
How do I install LLM X for offline use?
Visit https://mrdjohnson.github.io/llm-x/ in your browser. In your browser search bar, you should see a download/install button - press that to install it as a Progressive Web App. After installation, you can go offline and start chatting. You still need Ollama or another backend service running locally.