
This workflow, built on Flux, allows for one-click transformation of photos into personalized manga-style images and avatars. It combines LoRA for style control and ControlNet for content guidance, making it as simple as uploading a clear everyday photo—no parameter settings or prompt writing needed. For optimal results, it is recommended to use with LoRA models such as "FLUX | Classic Shojo Manga Style - Cassiopeia_TypeA" and "FLUX | Y2K Shojo Manga Style - Cassiopeia_TypeB." The output features bright colors, smooth linework, natural shapes, and expressive details, making it an excellent choice for character portraits!
Features:
-
Automatic Reverse Content Mapping: No need to input prompt keywords.
-
Pre-set LoRA Trigger Words: The includes default trigger words for "FLUX | Classic Shojo Manga Style - Cassiopeia_TypeA." If using a different LoRA, you can set your own trigger words.
-
Image Dimension Settings: Choose from three sizing methods (activate the desired node group); by default, the generated image size matches the reference image dimensions.
·Direct Reference Image Sizing: Automatically uses the dimensions of the reference image
·Enlarge Image Size: Enlarges the input to match the aspect ratio of the reference image. (Note: If the reference image is low-quality, it may affect ControlNet's ability to process it accurately.) You can also use this setting to upscale and enhance the reference image.
·Set Fixed Size and Crop: Useful for generating multiple images with the same size and similar composition. This option allows you to set the cropping area based on X and Y coordinates from the top-left corner of the image.
-
Nagtive Prompt Pre-Configuration: (To ensure style consistency, reverse prompts are pre-configured within the workflow. No adjustments are needed unless there are specific requirements
Instructions:
-
Upload a real-life photo of the person (a clear everyday photo is sufficient).
-
Select the desired LoRA model and enter the trigger words (Recommended LoRA models: "FLUX | Classic Shojo Manga Style - Cassiopeia_TypeA" and "FLUX | Y2K Shojo Manga Style - Cassiopeia_TypeB").
-
Click "Generate" to create the image.
❤️ Recommended Parameters ❤️ (Preset within the workflow; adjustments are unnecessary unless specific changes are needed)
-
Base Model: Flux
-
Image Size: 1024x1536 (Portrait orientation yields better character results)
-
Steps: 25–30
-
Sampler: DPM++2M
-
LoRA Weight: 0.8–1.0
-
ControlNet Strength: 0.2–0.3(higher values may impact LoRA effect)
-
ControlNet Start Percent: 0
-
ControlNet End Percent: 0.357
The included models are for learning and reference only; please do not use them commercially without authorization!
FLUX|古早少女漫画风格Ancient girl manga style-Cassiopeia_TypeA,Please visit the homepage for more information.
FLUX|Y2K少女漫画风格Y2K girl manga style-Cassiopeia_TypeB,Please visit the homepage for more information.
Thank you for the likes, downloads, and follows! Please feel free to share your creations and leave comments—thank you so much!
描述:
The current version is v1.0, and we will continue to optimize it based on any issues that arise and to make it more user-friendly.
训练词语:
名称: fluxOneClickPhotoTo_v10.zip
大小 (KB): 6
类型: Archive
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success