Z-IMAGE TURBO MANUAL INSTALL UPDATED

Manual Install Method from my last Youtube video:  https://youtu.be/apd68jTrxYc
AND NEW UPDATED VIDEO HERE: https://youtu.be/eyQkAW4wL1E


Manual Installation Guide for ComfyUI + Z-IMAGE TURBO
This guide explains how to manually install ComfyUI with the Z-IMAGE TURBO model, custom nodes and all required files.
It is split in two parts:



====================================================
PART A - LOCAL WINDOWS INSTALL (NVIDIA, WINDOWS ONLY)
STEP 1: PREREQUISITES
Before you start, make sure you have:


7-Zip (to extract ComfyUI)
Download: https://www.7-zip.org/


Git for Windows (to install custom nodes)
Download: https://git-scm.com/download/win


Windows 10 and 11 already include "curl", which we will use to download some files.
STEP 2: DOWNLOAD AND EXTRACT COMFYUI


Download the ComfyUI portable version for NVIDIA GPUs:
ComfyUI portable v0.3.75:
https://github.com/comfyanonymous/ComfyUI/releases/download/v0.3.75/ComfyUI_windows_portable_nvidia.7z


Once the file is downloaded, right click on it and extract it with 7-Zip.


After extraction, your folder should look like this:
ComfyUI_windows_portable
ComfyUI
models
custom_nodes
python_embeded
run_nvidia_gpu.bat
You will work inside this "ComfyUI_windows_portable" folder for the rest of the install.
STEP 3: INSTALL REQUIRED CUSTOM NODES


Go to the "custom_nodes" folder:
ComfyUI_windows_portable\ComfyUI\custom_nodes


In the Explorer window, click on the address bar at the top, type:
cmd
and press ENTER.
This opens a Command Prompt already inside the "custom_nodes" folder.


In this Command Prompt, run these commands one by one to clone all required nodes:
git clone https://github.com/ltdrdata/ComfyUI-Manager.git
git clone https://github.com/city96/ComfyUI-GGUF
git clone https://github.com/rgthree/rgthree-comfy
git clone https://github.com/yolain/ComfyUI-Easy-Use
git clone https://github.com/kijai/ComfyUI-KJNodes
git clone https://github.com/ssitu/ComfyUI_UltimateSDUpscale
git clone https://github.com/cubiq/ComfyUI_essentials
git clone https://github.com/wallish77/wlsh_nodes
git clone https://github.com/vrgamegirl19/comfyui-vrgamedevgirl
git clone https://github.com/ClownsharkBatwing/RES4LYF
git clone https://github.com/Jonseed/ComfyUI-Detail-Daemon
git clone https://github.com/ChangeTheConstants/SeedVarianceEnhancer
git clone https://github.com/Fannovel16/comfyui_controlnet_aux


Now you need to install the Python requirements for each node.
Go back to the root ComfyUI folder:
ComfyUI_windows_portable\
The portable version already includes Python inside "python_embeded", so you do NOT need to install Python yourself.
In this folder, open a new Command Prompt:


Click in the address bar


Type:  cmd


Press ENTER


Then run the following commands:
python_embeded\python.exe -m pip install -r ComfyUI\custom_nodes\ComfyUI-Manager\requirements.txt
python_embeded\python.exe -m pip install -r ComfyUI\custom_nodes\ComfyUI-GGUF\requirements.txt
python_embeded\python.exe -m pip install -r ComfyUI\custom_nodes\rgthree-comfy\requirements.txt
python_embeded\python.exe -m pip install -r ComfyUI\custom_nodes\ComfyUI-Easy-Use\requirements.txt
python_embeded\python.exe -m pip install -r ComfyUI\custom_nodes\ComfyUI-KJNodes\requirements.txt
python_embeded\python.exe -m pip install -r ComfyUI\custom_nodes\ComfyUI_UltimateSDUpscale\requirements.txt
python_embeded\python.exe -m pip install -r ComfyUI\custom_nodes\ComfyUI_essentials\requirements.txt
python_embeded\python.exe -m pip install -r ComfyUI\custom_nodes\wlsh_nodes\requirements.txt
python_embeded\python.exe -m pip install -r ComfyUI\custom_nodes\comfyui-vrgamedevgirl\requirements.txt
python_embeded\python.exe -m pip install -r ComfyUI\custom_nodes\RES4LYF\requirements.txt
python_embeded\python.exe -m pip install -r ComfyUI\custom_nodes\ComfyUI-Detail-Daemon\requirements.txt
python_embeded\python.exe -m pip install -r ComfyUI\custom_nodes\SeedVarianceEnhancer\requirements.txt
python_embeded\python.exe -m pip install -r ComfyUI\custom_nodes\comfyui_controlnet_aux\requirements.txt

If one of the folders does not have a "requirements.txt" file, you can skip that one.
STEP 4: DOWNLOAD Z-IMAGE TURBO MODEL FILES (LOCAL)
All Z-IMAGE TURBO files are hosted on Hugging Face under this base folder:
Base URL:
https://huggingface.co/Aitrepreneur/FLX/tree/main
You can either download from your browser or use "curl" in Command Prompt.
All paths below are relative to:
ComfyUI_windows_portable\ComfyUI\models\
4.1 TEXT ENCODER
Path:
ComfyUI_windows_portable\ComfyUI\models\text_encoders\
File:
Qwen3-4B-UD-Q6_K_XL.gguf
Create the "text_encoders" folder if it does not exist.
Example using curl (from the models folder):


Open Command Prompt in:
ComfyUI_windows_portable\ComfyUI\models


Run:
curl -L -o "text_encoders\Qwen3-4B-UD-Q6_K_XL.gguf" "https://huggingface.co/Aitrepreneur/FLX/resolve/main/Qwen3-4B-UD-Q6_K_XL.gguf?download=true"


4.2 VAE
Path:
ComfyUI_windows_portable\ComfyUI\models\vae\
File:
ae.safetensors
Example using curl (from the models folder):
curl -L -o "vae\ae.safetensors" "https://huggingface.co/Aitrepreneur/FLX/resolve/main/ae.safetensors?download=true"
4.3 UNET (MAIN Z-IMAGE TURBO MODEL)
Path:
ComfyUI_windows_portable\ComfyUI\models\unet\
Create the "unet" folder if it does not exist.
Choose ONE version based on your GPU VRAM:


For GPUs with less than 8 GB VRAM:
z_image_turbo-Q5_K_S.gguf


For GPUs with 8 to 12 GB VRAM:
z_image_turbo-Q6_K.gguf


For GPUs with 12 to 16 GB VRAM or more (best quality):
z_image_turbo-Q8_0.gguf


Example curl commands (run from ComfyUI_windows_portable\ComfyUI\models):
Q5_K_S (less than 8 GB VRAM):
curl -L -o "unet\z_image_turbo-Q5_K_S.gguf" "https://huggingface.co/Aitrepreneur/FLX/resolve/main/z_image_turbo-Q5_K_S.gguf?download=true"
Q6_K (8 to 12 GB VRAM):
curl -L -o "unet\z_image_turbo-Q6_K.gguf" "https://huggingface.co/Aitrepreneur/FLX/resolve/main/z_image_turbo-Q6_K.gguf?download=true"
Q8_0 (12+ GB VRAM, best quality):
curl -L -o "unet\z_image_turbo-Q8_0.gguf" "https://huggingface.co/Aitrepreneur/FLX/resolve/main/z_image_turbo-Q8_0.gguf?download=true"
Download only the version you plan to use.
4.4 UPSCALE MODELS
Path:
ComfyUI_windows_portable\ComfyUI\models\upscale_models\
Files:
4x-ClearRealityV1.pth
RealESRGAN_x4plus_anime_6B.pth
Example curl commands (from ComfyUI_windows_portable\ComfyUI\models):
curl -L -o "upscale_models\4x-ClearRealityV1.pth" "https://huggingface.co/Aitrepreneur/FLX/resolve/main/4x-ClearRealityV1.pth?download=true"
curl -L -o "upscale_models\RealESRGAN_x4plus_anime_6B.pth" "https://huggingface.co/Aitrepreneur/FLX/resolve/main/RealESRGAN_x4plus_anime_6B.pth?download=true"


4.5 MODEL PATCHES
Path:
ComfyUI_windows_portable\ComfyUI\models\model_patches\
Files:
Z-Image-Turbo-Fun-Controlnet-Union-fp8-e5m2.safetensors
Example curl commands:
curl -L -o "Z-Image-Turbo-Fun-Controlnet-Union-fp8-e5m2.safetensors" "https://huggingface.co/Aitrepreneur/FLX/resolve/main/Z-Image-Turbo-Fun-Controlnet-Union-fp8-e5m2.safetensors?download=true"


STEP 5: RUN COMFYUI WITH Z-IMAGE TURBO (LOCAL)


Go back to the main folder:
ComfyUI_windows_portable\


Double click:
run_nvidia_gpu.bat


Wait until the console finishes loading.
Then open your browser and go to:
http://127.0.0.1:8188/


You now have ComfyUI running locally with Z-IMAGE TURBO.



## **RunPod Installation**

1. Create account at https://bit.ly/runpodAi
2. Pick a 24 GB VRAM GPU with the **aitrepreneur/comfyui** template
3. Access port **8888**
4. Mirror the same folder structure from this guide
5. Use `curl` to grab files, e.g.:
   `curl -L -o "filename" "URL"`
6. Watch logs:
   `tail -f workspace/logs/comfyui.log`


You now have ComfyUI + Z-IMAGE TURBO running on RunPod.