This workflow works fine with 8GB VRAM, if you have a better card, you might want to disable the Lightning LoRA, add more steps, and cfg to every KSampler. Maybe 20 steps and cfg 2.5? You should try. Unlike the...
Laufzeitprofil
Quellenbeschreibung
Hello!
Version 2511 v1:
This workflow works fine with 8GB VRAM, if you have a better card, you might want to disable the Lightning LoRA, add more steps, and cfg to every KSampler. Maybe 20 steps and cfg 2.5? You should try. Unlike the previous versions of this workflow, this has a custom node and uses a Q version.
Resources:
[Model] Qwen-image-edit-2511-Q2_K.gguf :
[LoRA] Qwen-Image-Edit-2511-Lightning-4steps-V1.0-bf16
[CLIP] qwen_2.5_vl_7b_fp8_scaled
[VAE] qwen_image_vae
KI-generierter Kommentar
KI-generierte Erklärung auf Grundlage von Quellen- und Konfigurationsdetails. Vorschläge sind klar gekennzeichnet.
Qwen Image Edit Multi Gen is a Qwen workflow for editing a source image with text prompts and saving generated images.
It loads an image, encodes an edit prompt, samples, decodes the result, and saves it.
Use a source image and an edit prompt; examples include “orbit the camera 45 degrees,” “right-side,” and “zoom out.”
The output is a generated image saved with SaveImage.
Version 2511 v1 uses the custom node Comfyui-QwenEditUtils.
Listed model files: qwen_2.5_vl_7b_fp8_scaled.safetensors; qwen_image_edit_fp8_e4m3fn.safetensors; qwen_image_vae.safetensors; Qwen-Image-Lightning-4steps-V1.0.safetensors.
Node types include CFGNorm, CLIPLoader, ImageScaleToTotalPixels, KSampler, LoadImage, LoraLoaderModelOnly, and ModelSamplingAuraFlow.
Other node types are SaveImage, TextEncodeQwenImageEdit, UNETLoader, VAEDecode, VAEEncode, and VAELoader.
The description says it works with 8GB VRAM.
With a better graphics card, try disabling the Lightning LoRA, adding more steps, and setting cfg on every KSampler; 20 steps and cfg 2.5 are suggested examples.
Vorschlag · nicht geprüft
Adapt the example prompts to the original image; Simplified Chinese is recommended, and English is also an option.
Vorschlag · nicht geprüft
Before using it, check that Comfyui-QwenEditUtils and the listed node types are available in your ComfyUI setup.
Vorschlag · nicht geprüft
Before running, verify the model filenames: the description names Qwen-image-edit-2511-Q2_K.gguf and Qwen-Image-Edit-2511-Lightning-4steps-V1.0-bf16, while the listed model files are qwen_image_edit_fp8_e4m3fn.safetensors and Qwen-Image-Lightning-4steps-V1.0.safetensors.
Vorschlag · nicht geprüft
Muss das bearbeitet werden?
Melden Sie sich an, um eine Änderungsanfrage zu senden.
Quellen
1 QuelleQuellenauszüge
2 AuszügeSource context: 13936 downloads · Type Workflows · Base model Qwen
[Nodes] Comfyui-QwenEditUtils
About the prompts:
These prompts are just examples, so you might want to change them, because, depending on the original image they may work or may not work. Also, I totally recommend translating the prompts to Simplified Chinese, because in that language they will have more adherence. For example: First prompt is orbit the camera 45 degrees. Second prompt is right-side. Third prompt is zoom out. You can use English too, but, you'll need better luck.
Previous version text:
Hello!
People asked me to update this workflow, so, here it is! This version uses:
qwen_image_edit_2509_fp8_e4m3fn.safetensors
qwen_2.5_vl_7b_fp8_scaled.safetensors
And 8 steps LoRA:
Qwen-Image-Lightning-8steps-V2.0-bf16.safetensors
Also, I added a secondary image, if you want to add something to your new images, if you don't want, you can disable the load node. Remember always to change the prompts to do what you need.
The file you'll download has the JSON file as well as a PNG, so you can use either in ComfyUI
Cya!
Previous version (1.0):
Hello!
I'm kinda new to ComfyUI, so I made this spaghetti nightmare, this will create 6 different images from one image using Qwen Image Edit with 4 steps each one (I'm low VRAM!).
Files used:
Model : qwen_image_edit_fp8_e4m3fn
Lora : Qwen-Image-Lightning-4steps-V1.0
Clip : qwen_2.5_vl_7b_fp8_scaled
Vae : qwen_image_vae
Have fun!
Geschätzter VRAM Bedarf
Schätzung nicht verfügbar
Die Modell-Dateigrößen sind noch nicht vollständig genug, um diese Anforderung zu berechnen.
Anforderungen
17 AnforderungenCLIPLoader
MöglichText encoder · Unknown
qwen_2.5_vl_7b_fp8_scaled.safetensors
Datei ungeprüftText encoder · Hugging Face · Comfy-Org/Qwen-Image_ComfyUI
qwen_image_edit_fp8_e4m3fn.safetensors
Datei ungeprüftCheckpoint · Hugging Face · Comfy-Org/Qwen-Image-Edit_ComfyUI · Comfy-Org/Qwen-Image-Edit_ComfyUI · Hugging Face
VAE · Hugging Face · Comfy-Org/Qwen-Image_ComfyUI
Qwen-Image-Lightning-4steps-V1.0.safetensors
Datei ungeprüftLoRA · Hugging Face · lightx2v/Qwen-Image-Lightning
VAEDecode
MöglichVAE · Unknown
VAEEncode
MöglichVAE · Unknown
VAELoader
MöglichVAE · Unknown
CFGNorm
Nicht aufgelöstKnotenpaket · Unknown
ImageScaleToTotalPixels
MöglichKnotenpaket · Unknown
KSampler
MöglichKnotenpaket · Unknown
LoadImage
Source context: 13840 downloads · Type Workflows · Base model Qwen
Knotenpaket · Unknown
LoraLoaderModelOnly
MöglichKnotenpaket · Unknown
Knotenpaket · Unknown
SaveImage
MöglichKnotenpaket · Unknown
TextEncodeQwenImageEdit
Nicht aufgelöstKnotenpaket · Unknown
UNETLoader
MöglichKnotenpaket · Unknown