LOLLMS (Lord of Large Language and Multi modal Systems) is a free, open-source web interface that puts hundreds of AI models and personalities behind one friendly front end on your own machine. Created by ParisNeo, the project has just under 4,800 stars on GitHub and remains in active development in 2026, with a loyal community around its local-first approach to chatting with AI.
Rather than juggling separate apps for every model provider, LOLLMS installs once and connects to local model runners and cloud APIs alike, including Ollama, llama.cpp-style backends, OpenAI, Anthropic and many others. On top of chat it layers a system of switchable AI personalities, extensions for tasks like image generation and document discussion, and smart features such as long-term memory and multi-model discussions, all running from a local web UI you control. The personalities system is the project’s trademark: each persona carries its own instructions, model preferences and greeting style, and the community shares hundreds of them, from pair programmers and research assistants to dungeon masters for text adventures.
Free and open source under the Apache 2.0 license. Running local models costs nothing; if you connect paid cloud APIs, you pay those providers directly for your own usage.
AI enthusiasts, privacy-conscious users and tinkerers who want a single, extensible playground for many models without subscriptions or sending every conversation to a third-party app.