This is a private version trained on larger training parameters. Compared with the beta version, it delivers better performance in detail rendering, generalization and high-resolution output. This version is available only to my supporters. Resale is strictly prohibited. Please let me know if you find anyone selling it.

Preface

The evolution of AI is incredibly fast. No sooner had ✨MoreReal for Krea2Edit✨ been released than we got Qwen-Image-2.1 (QI-21 for short). QI-21 was trained at 2K resolution (while the comparable Edit model Klein was trained at 1K resolution), which means it can generate much more detailed visuals, so I couldn't wait to train the new MoreReal.

✨MoreReal for QI-21✨

It also comes in two versions: a beta version and a private version. The beta version is trained with a standard resolution (768px), while the private version uses a higher resolution (1536px), which perfectly matches the high-resolution feature of QI-21.

Parameters:

Sampler : euler/res_multistep

Scheduler : beta/sgm_uniform

Steps : 30/40

Config: 1/2

Trigger words : photorealistic

Exclusive Workflow :

QI-21 Single Image Editing Workflow

Solve the greasy image issue of QI-21

The scaling workflow of QI-21 supports a maximum of 32K

To maximize the performance of LoRA, it is recommended to set the resolution in the Text Encode Qwen Image 2.1 node to a value above 1536, or set it to 0 (to keep it consistent with the original image).

New Features

This is the biggest change in this version: it no longer generates generic AI faces (and of course, the faces it creates aren't as flawless as typical AI faces), and every single face it produces is completely unique.

It has a stronger ability to inherit the features in the input image, such as the exaggerated expressions in the figure.

It has a stronger perceptual ability for the input image; for example, it can recognize that the clouds and flowers in the image are not real.

The capability of rendering realistic backgrounds has been improved.

For exaggerated cartoon characters, it tends to generate toys more often.

but offers better support for characters

The handling of rust stains and bloodstains is rendered with greater realism and meticulous detail.

✨Member Event✨:

This LoRA is exclusive to LV3/LV4 users. Users who are not LV3/LV4 need not be disappointed. I will randomly select members every week to give away the private version of this LoRA. Whether you are an LV1 or LV2 user, leave a comment "✨MoreReal for QI-21✨" under this post now, and you will stand a chance to receive this gift. Hopefully you will be the next lucky winner.

OK! That's all! I hope you like it!

If you enjoy my sharing, you are welcome to join my private club. If you are already a member, you can get the project files for this sharing directly below. 👇👇👇

🫡May the AI-power be with you, see you soon!🫡

Disclaimer: My model or workflow is solely for learning and communication purposes and shall not be used for any commercial activities. All responsibilities for the generated content, including its legality, ethical implications, or potential misuse, are borne entirely by the user. By using this model or workflow, you agree to this principle. Please use it responsibly within the scope of local laws.

ESOZQa33WzFo81JhVACoD6VspU74pxVsKtsIpfbrZws

Click here