Deploying LibreChat Open-Source AI Chat Platform on Ubuntu 24.04

LibreChat is an open-source, ChatGPT-style web UI that supports OpenAI, Anthropic, Azure OpenAI, Gemini, OpenRouter, local OpenAI-compatible endpoints, and more — with MongoDB-backed conversation history and Meilisearch-powered search. This guide deploys LibreChat using its official Compose manifest plus a Traefik override for automatic HTTPS, following self-hosted AI chat platform deployment practices documented in Vultr Docs.

Clone LibreChat and Prepare the Environment

1. Clone the LibreChat repository and check out a stable tag:

git clone https://github.com/danny-avila/LibreChat.git
cd LibreChat
git checkout tags/v0.8.3

2. Find the Meilisearch data directory name pinned by this release:

grep -o 'meili_data_v[0-9.]*' docker-compose.yml | head -1

3. Create the required data directories (replace meili_data_v1.35.1 if the previous command printed a different name):

mkdir -p data-node images logs meili_data_v1.35.1 uploads
sudo chown -R 1000:1000 meili_data_v1.35.1

4. Copy the env template and uncomment the UID/GID lines:

cp .env.example .env
nano .env

UID=1000
GID=1000

Override the Compose Stack with Traefik

1. Create a Compose override that adds Traefik and wires the API to it:

nano docker-compose.override.yml

services:
  api:
    labels:
      - "traefik.enable=true"
      - "traefik.http.routers.librechat.rule=Host(`librechat.example.com`)"
      - "traefik.http.routers.librechat.entrypoints=websecure"
      - "traefik.http.routers.librechat.tls.certresolver=leresolver"
      - "traefik.http.services.librechat.loadbalancer.server.port=3080"
    volumes:
      - ./librechat.yaml:/app/librechat.yaml

  traefik:
    image: traefik:v3.6.10
    ports:
      - "80:80"
      - "443:443"
    volumes:
      - "/var/run/docker.sock:/var/run/docker.sock:ro"
      - "./letsencrypt:/letsencrypt"
    command:
      - "--providers.docker=true"
      - "--providers.docker.exposedbydefault=false"
      - "--entrypoints.web.address=:80"
      - "--entrypoints.websecure.address=:443"
      - "--entrypoints.web.http.redirections.entrypoint.to=websecure"
      - "--entrypoints.web.http.redirections.entrypoint.scheme=https"
      - "--entrypoints.web.http.redirections.entrypoint.permanent=true"
      - "--certificatesresolvers.leresolver.acme.httpchallenge=true"
      - "--certificatesresolvers.leresolver.acme.httpchallenge.entrypoint=web"
      - "--certificatesresolvers.leresolver.acme.email=admin@example.com"
      - "--certificatesresolvers.leresolver.acme.storage=/letsencrypt/acme.json"

2. Create an empty LibreChat configuration file (mounted by the override):

touch librechat.yaml

Start the Stack

1. Bring everything up:

docker compose up -d

Compose starts the LibreChat API, MongoDB, Meilisearch, and Traefik. Traefik requests a TLS certificate from Let's Encrypt over the HTTP-01 challenge and routes HTTPS to LibreChat on port 3080.

2. Verify the containers are running:

docker compose ps

3. Tail the logs:

docker compose logs --tail=50

You should see MongoDB initialize, Meilisearch start, Traefik register the route, and the LibreChat API listening on port 3080.

Register and Sign In

1. Open the registration page in a browser:

https://librechat.example.com/register

2. Fill in your name, username, email, and password and click Continue.

3. Sign in with the new credentials — the dashboard loads with the model selector and conversation interface.

To enable Google / GitHub / Discord / OpenID Connect sign-up, set ALLOW_SOCIAL_LOGIN=true and ALLOW_SOCIAL_REGISTRATION=true in .env and restart the stack.

Next Steps

LibreChat is running and served securely over HTTPS. From here you can:

• Plug in API keys for OpenAI, Anthropic, Azure, Gemini, or OpenRouter in .env

• Point LibreChat at a local OpenAI-compatible endpoint (Ollama, LocalAI) via librechat.yaml

• Enable file uploads, RAG, and image generation through the dashboard settings

For the full guide with additional tips, visit the original article on Vultr Docs.