A self-hosted web UI for your language models, written in Python.
Talks to anything that speaks the OpenAI API. Themed after Middle-earth.

---

LLeMbas is a server-rendered web application for chatting with language models you run yourself or reach over an API: chats in folders, a library of documents, notes and skills with search, tools and web search, agent chats that run on your own machines through the LLeMbas CLI, accounts with credits and usage, and an OpenAI-compatible API of its own. It is Python and FastAPI with htmx — no bundler, no build step, nothing fetched from a CDN at runtime — and keeps everything in one SQLite database.

Quick start

With Docker. The compose file builds the image locally from the checkout:

git clone https://github.com/LLeMbas/LLeMbas.git && cd LLeMbas
echo "LEMBAS_SECRET_KEY=$(python3 -c 'import secrets;print(secrets.token_urlsafe(48))')" > .env
docker compose up -d --build

It listens on 127.0.0.1:8080. Open it, create the first account — it becomes the administrator — and add a connection to your models. Keep the secret key: changing it signs everyone out and makes stored API keys unreadable. Other settings are in .env.example.

Put a TLS reverse proxy in front for anything beyond your own machine: the service worker and the microphone need HTTPS.

Updating:

git pull && docker compose up -d --build

On a machine of its own (systemd and nginx, or a Proxmox container): see deploy/README.md and https://llembas.eu/server.html.

Links

• Documentation: https://llembas.eu/docs/
• Changelog: https://llembas.eu/changelog.html
• The CLI: https://github.com/LLeMbas/LLeMbas-CLI

Licence

MIT.

A note on the theme

This is an independent project, themed as an affectionate nod to J.R.R. Tolkien's world. It is not affiliated with, endorsed by, or connected to the Tolkien Estate, Middle-earth Enterprises, or any related rights holder. All artwork here is original.