Real-Qwen-Image v2.0
本模型为 Qwen-Image-2512 的微调模型，比官方模型具有更好的锐度和真实感，以及对亚洲人脸美学进行了的优化。具体效果参见如下示例图片，本模型易于使用、参数适配性和LoRA兼容性良好，文字和出图品质俱佳，真正的开源生产力工具。
This model is a fine-tuned version of Qwen-Image-2512, have better sharpness and realism than the official model, and it has been optimized for the aesthetics of Asian faces. For specific effects, see the sample images below. This model is easy to use, has good parameter adaptability and LoRA compatibility, with excellent text and image quality, This model is a truly open-source productivity tool.
使用方法：
basic workflow pls refer to: Image
model shift: 1.0 - 8.0；
inference cfg: 1.0 -4.0；
inference steps: 10 -50；
sampler / scheduler: euler / simple ，or any other。
Also on: Huggingface.co, Modelscope
致谢：
感谢 kanttouchthis 提供量化脚本，sdnq_uint4_r128 量化版本模型，出图品质能达到 Q6 以上，接近 Q8，欢迎大家试用，需要安装 ComfyUI-SDNQ 插件。
Much thanks to kanttouchthis for providing the quantization script and the sdnq_uint4_r128 quantized model. The image quality can reach above Q6, nearly Q8. Everyone is welcome to try it; the ComfyUI-SDNQ plugin needs to be installed.
===============================================
Real-Qwen-Image v1.0 version:
本模型为 Qwen_Image 微调模型，主要提升了出图的清晰度和写实感。具体效果参见示例图片，图片中也附带有 ComfyUI 工作流，本模型极易使用、快速出图、LoRA兼容性良好。
The model is the Qwen_Image fine-tuned model, It enhances the clarity and realism of the generated images. For specific effects, please refer to the example images, which also include the ComfyUI workflow. The model is very easy to use and quickly generates images, and have a good LoRA compatibility.
For bf16 / gguf version you can download from my: https://www.modelscope.cn/models/wikeeyang/Real-Qwen-Image
模型使用：
基本组合：euler+simple，cfg 1.0，setps 20 -30，您可以尝试不同的组合。
Basic: euler+simple, cfg 1.0, setps 20 - 30, You can try more different combinations.
工作流 (Example Workflow) : Pls refer to - https://civarchive.com/images/97293023
Text_encoder model file link: https://huggingface.co/Comfy-Org/Qwen-Image_ComfyUI/tree/main/split_files/text_encoders, bf16/fp16/fp8/fp8_scaled/gguf version all ok.
Also on Huggingface.co
Description
本微调模型主要提升了 Qwen-Image 模型出图的清晰度和写实感。
This fine-tuned model enhances the clarity and realism of the generated images.
FAQ
Comments (23)
ohh yeah babyyyy lets go
Lets go......😂
Qwen creates not sharp (not actually blurry) images - how to mitigate that problem with some workflow tricks? Any working ideas???
Personally, maybe, I feel that it is still an issue with the model structure design. The Qwen model seems have less control layers for image color and clarity compare with Flux.
@wikeeyang too bad :(
@wikeeyang control layers? hahah what
how to fix error bro
did you use which workflow? I don't see the clip node from you posted image, then I guest maybe: 1. can't find qwen-vl text_encoder model, pls check model name and path; 2. need update ComfyUI core.
Have you fix it yet?
Don't pull clip link from node "Load checkpoint".
Place separate nodes to your workflow: "Load clip" for clip link to Lora or prompts, and probably VAE (Load VAE node, I can't see).
And download for it qwen-vl text_encoder model. (And maybe VAE).
閉源的都是lil bros
Yeah! Support Open Source...✌️
hello do you know where can I this QwenImgBlocksBuster+ ? it is missing on the workflow
thanks in advanced :)
Sorry! my mistake, this is the node only for debug the model layer which wrote by myself. 🤣
Pls ref the simple workflow for gererate. https://civitai.com/images/96878480.
oh ok big thanks for sharing this :D !
This is my fav Qwen model now.
Thanks a lot! 🤝✌️
Im sorry is this fake? I go to that workflow link and download workflow.
Workflow demands me to have qwen_image.vae, and qwen 2.5 7b bf16 clip. Its impossible to find where to download it from.
The only qwen 2.5 7b bf16 clip is made of 3 PARTS. how am i suppose to load it in clip. Isnt this real-qwen-image suppose to have it all inside?? its checkpoint, but workflow demands diffusion model.
I download the checkpoing file file and file name is NOT what is written in workflow. 19GB from this civitai website. it suppose to be checkpoint, why workflow is about model.. so confusing
you can download it from ComfyUI offical repo on HF site, https://huggingface.co/Comfy-Org/Qwen-Image_ComfyUI/blob/main/split_files/text_encoders/qwen_2.5_vl_7b.safetensors, this is the bf16 version text_encoder of qwen-image, https://huggingface.co/Comfy-Org/Qwen-Image_ComfyUI/blob/main/split_files/vae/qwen_image_vae.safetensors, this is the VAE.
I definitely like this qwen checkpoint , thanx for sharing
Thanks a lot! 🤝🤝🤝
@wikeeyang the model is really excellent, the final picture needs to be improved with some upscaler in workflow but otherwise it is excellent
@DiabolicX Thank you, Yes, may be qwen-image-edit-2509 will better a little, and you can try my Magic-Wan-Image or Flux1-DedistilledMixTuned v3.0-Krea for post refiner or upscaler model.