Have a ComfyUI workflow? Get a readable Workflow Doc in your browser.Make a workflow readable.Try it now ↗
Here's an interesting little T2I/I2I tool for you guys...
Runtime profile
Source description
Here's an interesting little T2I/I2I tool for you guys...
This is a multi-model workflow for Qwen ImageEdit 2511 and Wan 2.2 T2V .
So, like... why? 😂
This workflow is surprisingly useful. I picked Qwen 2511 and Wan 2.2 very purposefully.
Qwen 2511 is highly responsive to both text prompts, edits, character positioning, camera positioning, character consistency, and perhaps most importantly, allows huge malleability of a reference image. 🧈
Wan 2.2 T2V is exceptional at detailing and already functions as a 2-model generator, with high noise for composition and low noise for detail. So basically, we're just replacing the Wan high noise model with Qwen to take advantage of both model's strong suits.
AI-generated commentary
AI-generated explanation based on source and configuration details. Suggestions are clearly labeled.
Multi-Model Workflow - Qwen 2511 + Wan 2.2 is for text-to-image and image-to-image creation with text prompts and optional reference images; Qwen ImageEdit 2511 sets the composition, Wan 2.2 adds detail, and the workflow can produce detailed images or multiple-style iterations.
Qwen handles composition, while Wan 2.2 handles detail.
Use a text prompt for text-to-image work or a reference image for image-to-image work; the Reference Image option can be turned on or off in either mode.
Starting from an existing image, the workflow can add detail or carry a character or scene composition through to Wan's detail stage.
It can create multiple-style iterations from one image; the description notes that LoRAs make this easier.
The workflow is published by DirkDigitaller on Civitai and is listed with Qwen as its base model.
Listed model files include ComicRealism01a_CE_QWEN_AIT3k.safetensors, Duality02_CE_QWEN_AIT3k.safetensors, facetest-000007.safetensors, ghibli_style_qwen_v3.safetensors, low_noise_model.safetensors, and pencil-sketch-anime-qwen.safetensors.
Other listed model files are qwen_2.5_vl_7b_fp8_scaled.safetensors, qwen_image_edit_2511_fp8_e4m3fn_scaled_lightning_comfyui_4steps_v1.0.safetensors, qwen_image_vae.safetensors, qwen_masterpieces_v3.safetensors, SmoothMixStyle_Low.safetensors, umt5_xxl_fp8_e4m3fn_scaled.safetens
Listed node packs are comfyui-easy-use, comfyui-kjnodes, comfyui-rmbg, RES4LYF, rgthree-comfy, and was-ns.
Use Qwen/Wan Specific Prompting to direct details to the relevant model, and use Wan NAG to tell Wan what to focus on.
Suggestion · not verified
Use Reactor FaceSwap to reinforce Wan's character consistency.
Suggestion · not verified
Use the workflow notes when adjusting settings toward more consistency or more divergence.
Suggestion · not verified
If you use SageAttention, turn it off globally and toggle Patch SageAttention in the workflow because Qwen does not like the global setting.
Suggestion · not verified
Before setup, verify the intended files for qwen_2.5_vl_7b_fp8_scaled.safetensors and wan2.2_t2v_low_noise_14B_fp8_scaled.safetensors.
Suggestion
Does this need to be edited?
Sign in to send an edit request.
Sources
1 sourceSource excerpts
1 excerptSource context: 256 downloads · Type Workflows · Base model Qwen
Qwen sets the composition up, Wan knocks it down. 🎳
This can be used in a number of ways. It can be used to detail a pre-existing image, it can be used to set-up character or scene composition with follow through to Wan detail, it can be used as a creative tool to generate multiple-style iterations of one image (this is made easier with LoRAs).
This workflow has a lot going on. I've set up an options panel to help turn on an off relevant features, as well as a lot of explanation in the workflow itself. To run you through some of the options:
Reference Image - can be turned on and off for either I2I or T2I
Qwen/Wan Specific Prompting - target only the relevant model with certain details
Wan NAG - tell Wan what to focus on
Patch SageAttention - if you're using Sage, Qwen does not like this and it needs to be turned off as a global setting and toggled in workflow
Reactor FaceSwap - to reinforce Wan's character consistency
I've also included notes in the workflow to help you achieve your desired result - how to modify settings to either achieve increased consistency or increased divergence.
This one is pretty cool, I use it a lot. Great if you're not sure what you want and want to play with both minimalistic and high detail options. I think I'll make some variants with Pony/ZiT/SDXL as base also.
(Note - I use Qwen ImageEdit 2511 w/Lightning Lora embed, which means if you're using a different base model of Qwen ImageEdit you may want to download the 4-step LoRA to keep the steps/speed consistent.)
Enjoy! 😊🥳
Estimated VRAM requirement
Estimate unavailable
855 MB across 2 of 15 model files. Model file total + 25% loading overhead + 2 GB execution buffer, rounded up.
Requirements
21 requirementsComicRealism01a_CE_QWEN_AIT3k.safetensors
Not resolvedLoRA · Unknown
Duality02_CE_QWEN_AIT3k.safetensors
Not resolvedLoRA · Unknown
facetest-000007.safetensors
Not resolvedLoRA · Unknown
ghibli_style_qwen_v3.safetensors
Not resolvedLoRA · Unknown
low_noise_model.safetensors
Not resolvedLoRA · Unknown
pencil-sketch-anime-qwen.safetensors
Not resolvedLoRA · Unknown
qwen_2.5_vl_7b_fp8_scaled.safetensors
ConflictText encoder · Unknown
Possible sources
qwen_image_edit_2511_fp8_e4m3fn_scaled_lightning_comfyui_4steps_v1.0.safetensors
Not resolvedCheckpoint · Unknown
VAE · Unknown
LoRA · 563 MB · SAFETENSORS · Unknown
SmoothMixStyle_Low.safetensors
Not resolvedLoRA · Unknown
umt5_xxl_fp8_e4m3fn_scaled.safetensors
File unverifiedText encoder · SAFETENSORS · Unknown
VAE · Unknown
LoRA · 293 MB · SAFETENSORS · Unknown
wan2.2_t2v_low_noise_14B_fp8_scaled.safetensors
ConflictCheckpoint · Unknown
Possible sources
Node pack · Registry
Node pack · Registry
Node pack · Registry
Node pack · Registry
Node pack · Registry
was-ns
Not resolvedNode pack · Registry
