Self-hosted

Deploy VectorMBE in Seconds

One Docker image. Run it anywhere. Enter your license key on login to launch your own instance.

bash

docker run -d --name vectormbe \
  -p 8080:8080 -p 8081:8081 \
  radsilent/vectormbe:latest

Prefer a persistent setup? See Docker Compose, GPU, or platform guides.

Three steps

Pull. Run. License.

1. Pull the image

bash

docker pull radsilent/vectormbe:latest

2. Run

bash

docker run -d \
  --name vectormbe \
  -p 8080:8080 \
  -p 8081:8081 \
  radsilent/vectormbe:latest

Port 8080 serves the UI and API. Port 8081 is the WebSocket real-time sync channel.

3. Enter your license

Open http://localhost:8080 and enter your license key on login to launch your own instance. Each key activates one production instance.

Request License

Want to build from source or run the native desktop app? See the MBSE install guide.

Platform instructions

macOS, Windows, or Linux

The deploy package works anywhere Docker runs. Pick your platform for the exact commands.

Docker Desktop for Mac

Requires Docker Desktop (Apple Silicon or Intel).

bash · zsh

git clone https://github.com/radsilent/vectormbe-deploy.git vectormbe
cd vectormbe
docker compose up -d

Open http://localhost:8080.

Docker Desktop or WSL2

Requires Docker Desktop (WSL2 backend recommended) or WSL2 with Docker Engine.

PowerShell

git clone https://github.com/radsilent/vectormbe-deploy.git vectormbe
cd vectormbe
docker compose up -d

Or use WSL2 and run the Linux commands. Access at http://localhost:8080.

Docker Engine

Requires Docker Engine + Docker Compose.

bash

git clone https://github.com/radsilent/vectormbe-deploy.git vectormbe
cd vectormbe
docker compose up -d

Open http://localhost:8080.

Docker Compose

Persistent deployment

For production servers, use Docker Compose with auto-restart.

docker-compose.yml

services:
  vectormbe:
    image: radsilent/vectormbe:latest
    restart: unless-stopped
    ports:
      - "8080:8080"
      - "8081:8081"

Save as docker-compose.yml, then docker-compose up -d. Use docker compose up -d if you have the Compose plugin.

GPU acceleration

Enable PyTorch GPU inference

Docker Run

Requires the NVIDIA Container Toolkit.

bash

docker run -d \
  --name vectormbe \
  --gpus all \
  -e VECTORMBE_REQUIRE_TORCH_GPU=true \
  -p 8080:8080 \
  -p 8081:8081 \
  radsilent/vectormbe:latest

Docker Compose

Add the deploy block for GPU reservation.

docker-compose.yml

services:
  vectormbe:
    image: radsilent/vectormbe:latest
    restart: unless-stopped
    ports:
      - "8080:8080"
      - "8081:8081"
    environment:
      VECTORMBE_REQUIRE_TORCH_GPU: "true"
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: all
              capabilities: [gpu]

Requirements

Docker 24+ or any OCI-compatible runtime
1 CPU core, 512 MB RAM minimum (1 GB recommended)
Valid VectorMBE license key (entered on login to launch your own instance)
CPU inference works out of the box, no GPU required
Optional: NVIDIA GPU with Container Toolkit for accelerated embedding inference

Verify

Check that the container is healthy and the UI is reachable.

bash

docker ps --filter name=vectormbe
docker logs --tail 20 vectormbe
open http://localhost:8080  # macOS

On Linux use xdg-open http://localhost:8080; on Windows use start http://localhost:8080.