NQRust Analytics
Get Started

Installation

Deploy the full NQRust Analytics stack with the interactive installer — online, from a release, from source, or fully airgapped.

NQRust Analytics ships as a single interactive installer (nqrust-analytics) that deploys the entire platform with Docker Compose. This guide covers the prerequisites, the four installation methods, the installer screens, and the steps to perform once it finishes.

What gets installed

The installer deploys the complete stack as Docker containers:

ServiceDefault portRole
Analytics UI3000Web interface, user management, dashboards
Analytics Service5555AI service — text-to-SQL, answers, charts
Analytics Engine8080Core query processing engine
Ibis Server8000Python data-transformation layer
QdrantVector database for embeddings
Northwind DBPostgreSQL demo database available for immediate exploration

Authentication is enabled by default: the installer generates a secure JWT_SECRET, and a default administrator account is created so you can sign in immediately (see Sign in as the administrator).

Prerequisites

System requirements

  • A Linux host (Ubuntu 24.04 LTS or similar is recommended).
  • Permission to run Docker (run as root/sudo, or add your user to the docker group).
  • Approximately 15–40 GB of free disk space, depending on the installation method (the airgapped binary alone is ~3–4 GB).
  • A hostname or static IP reachable by the users who will access the application.

Software requirements

  • Docker Engine (engine + CLI) — install Docker.
  • Docker Compose v2 — invoked as docker compose (not the legacy docker-compose).
  • Docker Buildx (BuildKit) — usually bundled with Docker CE.

Two methods require additional tooling:

  • Build from source also requires the Rust toolchain and a GitHub Personal Access Token with the read:packages scope to pull images from GitHub Container Registry (ghcr.io).
  • Airgapped requires a build machine with internet access to produce the bundle; the target machine needs no connectivity.

Install

Select the method that fits your environment. Option A is recommended for most online installations.

Install the latest .deb and put nqrust-analytics on your PATH:

curl -fsSL https://raw.githubusercontent.com/NexusQuantum/installer-NQRust-Analytics/main/scripts/install/install.sh | bash

Then launch the installer:

nqrust-analytics install

Download the package

Download the latest .deb from the Releases page:

curl -LO https://github.com/NexusQuantum/installer-NQRust-Analytics/releases/latest/download/nqrust-analytics_*.deb

Install it

sudo apt install ./nqrust-analytics_*.deb
# or: sudo dpkg -i nqrust-analytics_*.deb

This adds the nqrust-analytics binary to /usr/bin.

Run the installer

nqrust-analytics install

Clone the repository

git clone https://github.com/NexusQuantum/installer-NQRust-Analytics.git
cd installer-NQRust-Analytics

Authenticate with GitHub Container Registry

docker login ghcr.io
# Username: your-github-username
# Password: your-personal-access-token (NOT your GitHub password)

The token must have the read:packages scope so Docker can pull the container images.

For environments with no internet access, the installer can be packaged as a single self-extracting binary (~3–4 GB) that bundles every Docker image.

Build the bundle (on a machine with internet)

git clone https://github.com/NexusQuantum/installer-NQRust-Analytics.git
cd installer-NQRust-Analytics
git checkout airgapped-single-binary

docker login ghcr.io
./scripts/airgapped/build-single-binary.sh

This produces nqrust-analytics-airgapped and a matching nqrust-analytics-airgapped.sha256 checksum file.

Transfer to the airgapped machine

Copy both files across via USB, SCP, or physical media:

cp nqrust-analytics-airgapped nqrust-analytics-airgapped.sha256 /media/usb/

Verify the checksum

On the target machine, confirm the binary arrived intact:

sha256sum -c nqrust-analytics-airgapped.sha256
# nqrust-analytics-airgapped: OK

If the checksum does not match, do not proceed — re-transfer the file.

Make it executable and run

chmod +x nqrust-analytics-airgapped
./nqrust-analytics-airgapped install

The first run auto-extracts the payload and loads all bundled Docker images, which can take several minutes.

If Docker itself is not installed on the airgapped host, use the offline Docker installer first — see docs/AIRGAPPED-INSTALLATION.md in the installer repo.

The interactive installer

Running nqrust-analytics install opens a terminal UI that guides you through configuration. Navigate with the arrow keys, select with Enter, and exit with Esc or Ctrl+C.

Generate the configuration

On the main screen, the installer checks whether a .env and a config.yaml already exist. On a fresh install they do not, so select Generate config.yaml to begin the guided setup.

The installer main screen showing the .env and config.yaml statusThe installer main screen showing the .env and config.yaml status

Choose an LLM provider

The installer lists the bundled provider templates. Select the one you want to use:

OpenAI, Anthropic, Azure OpenAI, Google (AI Studio & Vertex AI), AWS Bedrock, DeepSeek, xAI Grok, Groq, Qwen3, Zhipu, OpenRouter, Ollama, LM Studio, and a generic local-LLM template.

Selecting an LLM provider template from the installer listSelecting an LLM provider template from the installer list

Enter the provider API key

Provide the API key for the provider you selected. Local providers such as Ollama or LM Studio may not require a key.

Entering the API key for the selected providerEntering the API key for the selected provider

Save the configuration

Select Save from the menu (or Cancel to discard). The installer writes both .env and config.yaml — generating a secure JWT_SECRET for authentication in the process — and returns you to the main screen.

Proceed with installation

Now that .env and config.yaml exist, the main-screen option changes to Proceed with installation. Select it to begin. The installer streams the Docker Compose operations with per-service status; wait until it completes.

The main menu also offers Configure Identity SSO (optional) to enable single sign-on before you proceed — see Single Sign-On.

The installation progress screen with the per-service statusThe installation progress screen with the per-service status

Installation complete

On success, the installer displays the result and full logs. The platform is now running and ready to use.

Post-installation setup

Open the application

Open http://<server-host>:3000 in your browser, where <server-host> is your server's IP address or hostname.

If you are browsing from the server itself, http://localhost:3000 also works. To reach the application from other machines, use the server's IP address or hostname instead of localhost.

Activate the license

On first launch, the License Activation screen appears. Accept the End User License Agreement (EULA), enter the license key provided by the NQRust team, and select Activate License. For an offline deployment, upload a signed .lic file instead. You can review the license later under Settings → License.

The License Activation screen with the EULA agreement and the license key fieldThe License Activation screen with the EULA agreement and the license key field

Sign in as the administrator

A default administrator account is created during setup:

email:    admin@localhost
password: admin123
The Analytics UI sign-in screenThe Analytics UI sign-in screen

Change this password immediately after your first sign-in. Additional users are created by an administrator from the User Management page — see Accounts & Sign-in.

(Optional) Enable single sign-on

In addition to email/password, you can enable NQRust Identity (SSO) from the installer's Configure Identity SSO screen (protocol, host, port, realm, client ID/secret, default role, auto-register). To change it later, re-run the installer — update the SSO settings (or edit the KEYCLOAK_* variables in .env) — and select Proceed with installation to apply the change.

See Single Sign-On for the full form and variable reference.

Create your first dashboard

From the Home page, create a dashboard and ask questions in natural language against the demo data, or connect your own data source.

Invite your team

Share dashboards (view or edit) and chat threads with other users to collaborate.

To connect your own database instead of the demo, continue to the data source connection guides.

On this page