comfyui-nitra 1.0.25

Nitra custom node for ComfyUI

Usage License: Nitra Labs grants you a limited license to run the Nitra plugin inside ComfyUI for its intended functionality in commercial or non-commercial workflows. Any other commercial use—including copying, modifying, sublicensing, or redistributing this code or derivative works—requires prior written consent from Nitra Labs.

Nitra makes private AI generation in ComfyUI effortless — built for film and VFX artists who demand privacy, polish, and more time to create. A plug-in right in ComfyUI, which makes the newest, professional-level, workflows a click away.

About Nitra Labs

Nitra Labs is an agency that excels at implementing, scaling, and finetuning AI pipelines in the film and VFX industries. We focus on security and data because your content is your backbone, and uploading your content to any cloud AI service risks your content becoming training material in the future.

Table of Contents

• The Nitra Advantage
• Installing Nitra
  • Auto-Installers
  • Comfy Manager
  • Manual Install
  • RunPod Template
• Create an Account
  • Free Features
  • Paid Features
  • How to create an account
  • How to subscribe
• Using Nitra
  • ComfyUI Optimizer
  • Workflow Installer
    • Subgraphs
  • Model Installer
  • Install Missing (Coming Soon)
• User Configuration
• How Can We Help?
• Logout

The Nitra Advantage

The Nitra Plug-in for ComfyUI gives you instant access to premium workflows specifically designed for production pipelines. The Nitra difference is when you download a workflow, you get everything needed to run the workflow. No more searching for obscure custom nodes or models. On top of that, the workflows are condensed into easy-to-use subgraphs that can be searched just like any node. Let's walk through an example.

Let's install Z-Image-Turbo

After clicking Install, all workflows, subgraphs, custom nodes, and models for running Z-Image-Turbo will install.

Now let's restart ComfyUI, and check out what was installed for us!

This is the workflow that was downloaded. It contains a Z-Image-Turbo subgraph with documentation for what all of the inputs and outputs of the subgraph do. The models were already installed, so just click "Run" and as long as you have the VRAM to run the workflow, it will run immediately.

Here is the searchable subgraph. Search for it as you would any other node, and connect it into your favorite image-to-video workflow or anywhere else you'd like.

Nitra has created an extremely efficient pipeline for getting workflows into your hands quickly, so when a new model is released, Nitra is committed to delivering the capability to you quickly and seamlessly.

Installing Nitra

Nitra Labs offers a variety of ways to install the Nitra Plug-in into ComfyUI for optimal convenience.

Auto-Installers

Download the applicable script for your system, which can be found at this link.

Windows Install

Place the .bat file in the folder location on windows that you want to install ComfyUI-Nitra in. Double click on the .bat file, and it will start to install ComfyUI with Nitra.

Inside the ComfyUI Folder, you will find a file called "run_comfy.bat". Double click this file to launch ComfyUI. Wait until the terminal stops producing text and then go to the next step.

Navigate to localhost:8188 or 127.0.0.1:8188 in your web browser of choice (Firefox or Google Chrome recommended).

Linux Install

Place the .sh file in the folder location that you want to install ComfyUI-Nitra in. Open a shell and run:

./linux_install_nitra_comfy.sh

This will clone ComfyUI into your current folder. Inside the ComfyUI Folder, you will find a file called "run_comfy.sh". Double click this file or open a shell and run it to launch ComfyUI. Wait until the terminal stops producing text and then go to the next step

Navigate to localhost:8188 or 127.0.0.1:8188 in your web browser of choice (Firefox or Google Chrome recommended).

Comfy Manager

Manual Install (Wheel)

Download the platform-specific wheel from the dist/ folder (Windows or Linux) and install it directly into your ComfyUI/custom_nodes directory. From the root of your ComfyUI install:

cd ComfyUI/custom_nodes
pip install ../../dist/<platform>/comfyui_nitra-<version>-cp39-abi3-<platform>.whl -t ComfyUI-Nitra --upgrade --no-deps

After the install completes you should see a ComfyUI-Nitra folder containing the compiled binaries and static assets. Restart ComfyUI and refresh your browser to load the plugin.

RunPod Template

Use our referral code to sign up for RunPod if you don't already have an account! https://runpod.io?ref=rqu7tin8

Nitra Labs provides a RunPod template to quickly deploy ComfyUI from anywhere, even if you don't own a GPU yourself. You can find it at this link

When you click on the link and have a valid account created, you'll be prompted to start a pod.

Nitra Labs recommends setting up Network storage of 500gb minimum, preferably up to 1TB or more to handle the large size of the models. This will allow you to keep all your work and you won't have to restart all over again each time you start a new pod.

Select your network storage, and CUDA Versions as shown below. CUDA 12.8 or 12.9 are required becasue of blackwell architecture support for RTX 50xx series.

Then select a GPU from the list below. For most production work, Nitra Labs recommends an RTX 5090 at minimum. RTX 5090 is an excellent balance of speed and value.

Once you've selected your GPU, and confirm that the ComfyUI-Nitra template is selected, click "Deploy On-Demand"

If this is your first time launching ComfyUI on your network storage or if you do not plan to use a network storage, expect about 20mins of first time setup.

Go to the logs of your RunPod pod, and wait until you see the circled message in the container logs. That tells you that ComfyUI is ready.

Click "Connect", and then click "ComfyUI" to open ComfyUI. A JupyterLab session is also provided, where you can view the pod's filesystem.

Docker

The docker command to run the ComfyUI-Nitra container is:

docker run -d \
  --name comfy-nitra \
  --gpus all \
  -p 8188:8188 \
  -p 8888:8888 \
  -v /path/to/workspace:/workspace \
  theartofficial/comfyui-nitra:latest

Create an Account

Free Features

• PyTorch Version Switcher
• Sage Attention Installer
• Windows Triton Installer
• Slow Depth & Pose Map Fix
• Cuda Toolkit Installer
• Windows Build Tools Installer
• Launch Shell with Build Tools Activated

Paid Features

• Auto-Install 100+ top-quality curated and subgraphed workflows
• 400+ workflows in the roadmap
• Subgraphs published straight into your library to use as nodes
• Model Auto-Installer
• (coming soon) Install All Missing Models & Custom Nodes for a workflow

How to create a free account

Once you have Nitra installed into ComfyUI, click on the nitra button at the top of ComfyUI.

Click the "Continue with Login" button on the dialog that pops up. All authentication is handled by Auth0 for optimal user security. The only personal data that Nitra Labs stores are name and email address for account handling.

Authenticate the Auth0 form with either email/password, github, or google.

Once you've completed log-in, you should be re-directed back to ComfyUI, where you will see the Nitra Dashboard appear.

How to subscribe to Nitra Premium

To purchase a license to access over 100 premium workflows and subgraphs, click the "Purchase License" button on the left side.

That will bring you to our Nitra Account Dashboard, where you can Upgrade your account to the paid version.

All purchases are handled securely through Stripe. Nitra does not store any of your financial information, all financial transactions run through Stripe.

After subscribing, head back to ComfyUI, refresh your browser, and you should see your account upgraded to "Premium Subscription" on the left side.

Using Nitra

ComfyUI Optimizer

Workflow Installer

Subgraphs

Model Installer

Install Missing (Coming Soon!)

When Implemented, this will allow users to open any workflow in ComfyUI and Nitra will attempt to find all custom nodes and models needed for that workflow, and install them for you. No more searching through the manager or scouring huggingface to find the correct model.

User Configuration

How Can We Help?

Logout