Home Creators Posts Import Register Favorites Logout

Downloads

Content

Hello my friends, a new amazing image model is out:

Qwen-Image-2.1. It's an Image Edit model that can handle up to 10 input images. I'm working on an official YT video right now to show you everything you can accomplish with it.

And it handles real PNG transparency, which is great for logos and for further editing in general.

I found this to be the perfect opportunity to celebrate by giving every paying member access to my new V3 Easy Installer, like I promised in an earlier post. 🥳

If you are at a higher tier and already have the Installer, you should see an update.

If you don't have it yet, just download the zip file, unzip and run the setup.

As always it will handle ComfyUI, all the nodes, the model download and is even able to use your custom folders and will put the workflows in your ComfyUI folder.

You can find the workload in the ComfyUI section under Qwen-Image 2.1 .

I retired the old Qwen workloads but you can still show them but activating this toggle in the header filtering.

If you prefer the manual installation over my Installer, you need to install everything listed below. The workflows are attached to this post, but as usual they should also end up in your ComfyUI once the installation completes.

Here's everything you need to run Qwen-Image-2.1 in your own ComfyUI. Grab the files below, drop each one into the folder shown, and you're good to go.

🐍 Python & PyTorch

This workload was built and tested with:

Python 3.12

PyTorch 2.11.0 with CUDA 13.0

📦 ComfyUI & custom nodes

Start with ComfyUI itself:https://github.com/comfyanonymous/ComfyUI

Then install these custom nodes into ComfyUI\custom_nodes (or search for them by name in ComfyUI-Manager):

ComfyUI-IntoTheLatent-Utils

comfyui_controlnet_aux

ComfyUI-Manager

rgthree-comfy

cg-use-everywhere

🎨 Models

🖼️ Diffusion models → ComfyUI\models\diffusion_models

qwen_image_2.1 – pick the one that fits your graphics card's VRAM:

8 GB: qwen_image_2.1_int8_convrot.safetensors

24 GB: qwen_image_2.1_bf16.safetensors

📝 Text encoders → ComfyUI\models\text_encoders

qwen3vl_8b – pick the one that fits your graphics card's VRAM:

8 GB: qwen3vl_8b_int8_convrot.safetensors

24 GB: qwen3vl_8b_bf16.safetensors

🎭 VAE → ComfyUI\models\vae

qwen_image_2.1_vae_bf16

qwen_image_2.1_vae_bf16.safetensors

⬇️ Download the workflows below this post ⬇️

Files

Installer V3 - First Impression Walkhtrough

In this walkthrough I take you through the whole thing from first download to a finished ComfyUI install. You will see how to get past the browser and antivirus warnings that show up because the build is not code-signed, how the install wizard works, and how the app registers itself in your Windows installed apps so you can uninstall it cleanly. From there I show the update system in practice: app updates with a live download and a restart-and-install step, the choice between the stable and preview branches, and catalog updates that refresh the ComfyUI workloads without touching the installer itself. Then we install ComfyUI end to end — a blank install with the Manager, Triton and Sage Attention, or a preset that pulls image or video models, filtered by type. I also cover using your own model folders with custom subfolder names and your own mappings for folders ComfyUI does not create by default, plus the new step-by-step logging panel that shows exactly where the install is. I finish with a second parallel install alongside an existing one, so you can see how running multiple instances works. From here the next updates will be about workflows rather than the software, which I consider close to feature complete. Soon this installer will be available to all paying members but I wanted my most loyal supporters to be the first to get their hands on the New Installer.

Comments

No comments found for this post.