GTEdgeAI/gt-ai-os-community
3
0
Fork 0
Code Issues Packages Releases Wiki Activity
Page: Installation
Pages
Cloudflare Tunnel Setup Control Panel Guide Home How to build GT AI OS Agents Installation Ollama Setup Projects for NVIDIA NIMs and Nemotron using Local Ollama Tenant App Guide Troubleshooting Updating
Clone
3
Installation
daniel edited this page 2026-01-10 03:39:22 +00:00
Table of Contents

Installation

TLDR installation overview:

  1. Check your host operating system and processor type to ensure you have compatible hardware/virtual machine. Note that only Unix based operating systems are supported (Windows is not supported).

  2. Minimum system requirements: 4 CPU cores, 16GB RAM, 50GB SSD. The GT AI OS only needs about 7GB to run, it will check that the target system has at least 16GB at install time so that the installation will not crash the host.

  3. Follow the installation scripts as provided because all the steps must be run in the order as shown below. If you don't follow the provided process, the installation might/will likely fail.

  4. Once the installation is finished, first log into the Control Panel to configure your AI models and create your account:

Open your web browser Go to the Control Panel at: http://localhost:3001 (or http://hostip:3001 over a network)

Log in with the default admin credentials:

  • Email: gtadmin@test.com
  • Password: Test@123

In the Control Panel referred to as "CTP":

  • Add your API keys (NVIDIA NIM, Groq, or configure Ollama for local models. Detailed info is provided in the CTP.
  • Configure AI models (a number of models from NVIDIA and Groq are preloaded for your convenience)
  • Create your own admin account
  • Delete the default admin account (important for security)
  1. After configuring the Control Panel, log into the Tenant App with your new account:

Go to the Tenant App at: http://localhost:3002 (or http://hostip:3002 over a network)

Log in with the account you created in step 4 (not the default credentials).

  1. Note that there are separate URLs: the Control Panel (localhost:3001) is for configuration, and the Tenant App (localhost:3002) is your daily interface for working with AI agents.

  2. We have prebuilt Agents here: Demo agent template files

Inference sources must be configured in Control Panel before you upload and configure Agents from the configuration csv files.

You can download these csv files and upload them in the Agent Configuration menu in the Tenant.

This will give you some Agents to start working with. You can create your own agents anytime, connected to any valid inference source that you have configured in Control Panel and with whatever datasets you have configured in the Tenant.

Enjoy!

Please create an Issue if you have any problems at all with the installation documentation or anything else so we can fix it.

Table of Contents

Supported Platforms

Platform Host Architecture Status
Ubuntu Linux 24.04 x86_64 Supported
NVIDIA DGX OS 7 (Optimized for Grace Blackwell Architecture) ARM64 Supported

Prerequisites

  • Minimum RAM and storage per platform above
  • Internet connection for initial setup

Ubuntu Linux 24.04 (x86_64)

Step 1: Install Docker and Docker Compose

  1. Open a Terminal window
  2. Copy and paste this entire command block, then press Enter:
sudo apt update && sudo apt install -y curl git ca-certificates gnupg lsb-release && \
curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo gpg --dearmor -o /usr/share/keyrings/docker-archive-keyring.gpg && \
echo "deb [arch=amd64 signed-by=/usr/share/keyrings/docker-archive-keyring.gpg] https://download.docker.com/linux/ubuntu $(lsb_release -cs) stable" | sudo tee /etc/apt/sources.list.d/docker.list > /dev/null && \
sudo apt update && sudo apt install -y docker-ce docker-ce-cli containerd.io docker-buildx-plugin docker-compose-plugin && \
sudo usermod -aG docker $USER && \
sudo systemctl start docker && \
sudo systemctl enable docker

  1. Important: Log out of your computer completely, then log back in. This step is required for Docker permissions to take effect.

  2. Verify installation:

docker --version
docker compose version

Step 1b: (Optional) Enable NVIDIA GPU for Embeddings

If your system has an NVIDIA GPU and you want to use it for faster document embeddings:

1. Check if NVIDIA drivers are already installed:

sudo apt install nvtop && nvtop

If this command shows your GPU info, skip to step 3. If not, continue with step 2.

2. Install NVIDIA drivers:

Update package list

sudo apt update

Install the recommended NVIDIA driver

sudo ubuntu-drivers install

Reboot to load the new driver

sudo reboot now

After reboot, verify the driver is working:

nvtop

You should see your GPU model, driver version, and CUDA version.

3. Install NVIDIA Container Toolkit:

Add NVIDIA package repository

curl -fsSL https://nvidia.github.io/libnvidia-container/gpgkey | sudo gpg --dearmor -o /usr/share/keyrings/nvidia-container-toolkit-keyring.gpg
curl -s -L https://nvidia.github.io/libnvidia-container/stable/deb/nvidia-container-toolkit.list | \
  sed 's#deb https://#deb [signed-by=/usr/share/keyrings/nvidia-container-toolkit-keyring.gpg] https://#g' | \
  sudo tee /etc/apt/sources.list.d/nvidia-container-toolkit.list

Install the toolkit

sudo apt update
sudo apt install -y nvidia-container-toolkit

Configure Docker to use NVIDIA runtime

sudo nvidia-ctk runtime configure --runtime=docker
sudo systemctl restart docker

4. Verify Docker can see your GPU:

docker run --rm --gpus all nvidia/cuda:12.1.0-base-ubuntu22.04 nvidia-smi

You should see the same GPU info as running nvtop directly.

Note: GPU support is automatically detected during installation. If NVIDIA Container Toolkit is installed, the installer will enable GPU acceleration for embeddings. If not, it will use CPU mode (which works fine, just slower for large document processing).

Step 1c: Install Git

sudo apt install -y git

Step 2: Download and Install GT AI OS

  1. Open Terminal again and paste these commands one at a time:
git clone https://gitea-dell-promax-gb10.gtedgeai.app/GTEdgeAI/gt-ai-os-community.git
cd gt-ai-os-community
sudo ./installers/install-gt2-ubuntu.sh
  1. Wait for installation to complete (10-15 minutes)

Step 3: First-Time Setup (Control Panel)

Important: You must configure the Control Panel before using the Tenant App.

  1. Open your web browser and go to the Control Panel: http://localhost:3001

  2. Log in with the default admin credentials:

    • Email: gtadmin@test.com
    • Password: Test@123

  3. Complete initial setup (see Control Panel Guide for details):

    • Add your API keys (NVIDIA NIM, Groq, or set up Ollama for local models)
    • Initialize AI models
    • Create your own admin account
    • Delete the default admin account (important for security)

Step 4: Start Using GT AI OS (Tenant App)

  1. Open the Tenant App: http://localhost:3002
  2. Log in with your new account (the one you created in Step 3)
  3. Start building AI agents! See the Tenant App Guide for help.

Optional setup:

Having issues? Check the Troubleshooting guide.

NVIDIA DGX OS 7 (Grace Blackwell Architecture)

Docker and Docker Compose are already installed on DGX OS.

Step 1: Verify Docker is Available

docker --version
docker compose version

Step 2: Add Yourself to Docker Group

If you haven't already:

sudo usermod -aG docker $USER && newgrp docker

Step 3: Download and Install GT AI OS

git clone https://gitea-dell-promax-gb10.gtedgeai.app/GTEdgeAI/gt-ai-os-community.git
cd gt-ai-os-community
sudo ./installers/install-gt2-dgx.sh

Wait for installation to complete (10-15 minutes).

Step 4: First-Time Setup (Control Panel)

Important: You must configure the Control Panel before using the Tenant App.

  1. Open your web browser and go to the Control Panel: http://localhost:3001
  2. Log in with the default admin credentials:
    • Email: gtadmin@test.com
    • Password: Test@123
  3. Complete initial setup (see Control Panel Guide for details):
    • Add your API keys (NVIDIA NIM, Groq, or set up Ollama for local models)
    • Initialize AI models
    • Create your own admin account
    • Delete the default admin account (important for security)

Step 5: Start Using GT AI OS (Tenant App)

  1. Open the Tenant App: http://localhost:3002
  2. Log in with your new account (the one you created in Step 4)
  3. Start building AI agents! See the Tenant App Guide for help.

Optional setup:

Having issues? Check the Troubleshooting guide.