I was having so much trouble finding a working Qwen ImageEdit 2511 inpainting workflow, so I've made one. Every other workflow I tried to use seemed to be broken or not work effectively. This one is simple to use -
Profil d'exécution
Description de la source
I was having so much trouble finding a working Qwen ImageEdit 2511 inpainting workflow, so I've made one. Every other workflow I tried to use seemed to be broken or not work effectively. This one is simple to use -
Use ComfyUI Mask Editor to create your image mask.
In prompt, describe the image, then specify the changes you want to the masked area.
That's it. Easy. ImageEdit is so responsive to text prompts and editing with image consistency that you don't always need inpaint, but for targeted changes with multiple primary subjects in an image, this tool has proved priceless and forms a core tool in my set.
I've added the option to turn 'Reference Latent' on and off - it's very useful for image consistency with original, but can get in the way of more creative edits. Use at your discretion.
Commentaire généré par l’IA
Explication générée par l’IA à partir des détails de la source et de la configuration. Les suggestions sont clairement signalées.
Inpaint (Qwen ImageEdit 2511) is for targeted edits to an existing image: use a ComfyUI Mask Editor mask and a text prompt describing the change to create an edited image.
Create the mask in ComfyUI Mask Editor, describe the image and requested changes in a prompt, and use both for the masked edit.
Use an existing image and a mask marking the area you want to change.
Write a prompt that describes the image and specifies the changes for the masked area.
The workflow is intended to produce an edited image with changes applied to the masked area.
This is a Qwen-based workflow published by DirkDigitaller.
The workflow lists qwen_2.5_vl_7b_fp8_scaled.safetensors, detailz_qwen_000024000.safetensors, QwenImageEditInpaint_v1.safetensors, and qwen_image_edit_2511_fp8_e4m3fn_scaled_lightning_comfyui_4steps_v1.0.safetensors.
It also lists aldniki_qwen_reality_transform_v01.safetensors, qwen_image_inpaint_diffsynth_controlnet.safetensors, and qwen_image_vae.safetensors.
The listed node packs are comfyui-kjnodes version 1.2.9 and comfyui_layerstyle version 2.0.38.
Use Reference Latent for closer consistency with the original; turn it off when you want more creative edits.
Suggestion · non vérifié
The description reports a four-step sampler with the author's Lightning LoRA setup; without Lightning LoRA or an embed, increase the sampler step count.
Suggestion · non vérifié
Before use, verify that the listed model files and node packs are available in your ComfyUI setup.
Suggestion · non vérifié
Check that qwen_image_vae.safetensors is the VAE file intended for your setup before using it.
Suggestion · non vérifié
Faut-il modifier cela ?
Connectez-vous pour envoyer une demande de modification.
Sources
1 sourceExtraits de sources
1 extraitSource context: 1298 downloads · Type Workflows · Base model Qwen
I use Qwen ImageEdit 2511 with Lightning Lora embed, but this should work fine with other versions of 2511. If you're not using lightning lora or embed, make sure you increase the number of steps in the sampler - I use 4-step with my model version.
This workflow takes advantage of the Qwen Image Edit Easy Inpaint Lora by _Envy.
Estimation des besoins VRAM
Estimation indisponible
Les tailles des fichiers du modèle ne sont pas encore suffisamment complètes pour calculer cette exigence.
Exigences
Exigences 9aldniki_qwen_reality_transform_v01.safetensors
Non résoluLoRA · Unknown
detailz_qwen_000024000.safetensors
Non résoluLoRA · Unknown
qwen_2.5_vl_7b_fp8_scaled.safetensors
ConflitText encoder · Unknown
Sources possibles
qwen_image_edit_2511_fp8_e4m3fn_scaled_lightning_comfyui_4steps_v1.0.safetensors
Non résoluCheckpoint · Unknown
qwen_image_inpaint_diffsynth_controlnet.safetensors
Non résoluCheckpoint · Unknown
VAE · Unknown
QwenImageEditInpaint_v1.safetensors
Non résoluLoRA · Unknown
Pack de nœud · Registry
Pack de nœud · Registry