Leafy Chat

For everyone

Leafy Chat is a ChatGPT-style assistant that runs entirely on local hardware. Ask it questions, get help writing something, brainstorm, or flip on web search so it can pull in live results before answering. It can also generate images and render interactive artifacts right inside the conversation. Create an account to try it; new sign-ups wait for a quick admin approval before their first chat.

What you can do

• •Chat with a local AI model for questions, writing help, or brainstorming
• •Toggle web search to pull in live SearXNG results before it answers
• •Generate images directly from a text prompt
• •View and interact with code or rendered artifacts inline in the chat
• •Every prompt and response stays on the server, nothing is sent to an outside AI provider
• •Sign up for an account, then wait for admin approval before first use

For admins

Open WebUI sits in front of Ollama, which runs the language model on the server's NVIDIA GPU for fast local inference. Self-registration is on, but new accounts land in a pending role until manually approved, keeping it invite-only in practice. Web search queries route through SearXNG's RAG integration for live results, and both image generation and shareable chat artifacts are enabled.