
This model is trained based on the MajicFlux model ecosystem, focusing on Tyndall effect lighting and enhanced light-shadow contrast textures. It requires the Majicflus model for optimal performance, and using it with Flux_dve may not yield satisfactory results.
I personally use Forge for image generation, and users can customize their workflow with Majicflus as needed.
Technical Parameters:
-
CFG: Default 3.5
-
Sampling steps: Default 20, can be increased to 30 if needed
-
LoRA weight: 0.7
-
Additional face and hand refinement (AD) is recommended for portraits with smaller face/body proportions
Trigger word: dingdaer
Example prompt: A girl in black clothing stands by a dark wall, with light and shadow falling on half of her face, creating a strong contrast between light and dark. Light shines through, and smoke fills the air.,dingdaer
"MAJICFLUS_麦橘丁达尔" Model (hereinafter referred to as "the Model") is developed and released by [DonRat] (hereinafter referred to as "the Owner"), based on MajicFlux's FLUX-lora model.
License Terms:
The Owner grants individuals and organizations permission to use images generated by this model free of charge for non-commercial educational or informational purposes, provided that:
-
Users comply with relevant laws and do not infringe upon the legal rights of the model or any third parties
-
Attribution must be given as "Generated by [DonRat]'s 'MAJICFLUS_麦橘丁达尔' Model" when using the images
For commercial use, a commercial license agreement must be signed with the Owner. For commercial licensing and custom model inquiries, please contact the Owner through their profile information.
The Owner will continue to provide free model updates to individual users as a gesture of support for community open-source contributors. Commercial user partnerships help drive the development and continuous improvement of this model. We appreciate every user's understanding and support.
描述:
This model is trained based on the MajicFlux model ecosystem, focusing on Tyndall effect lighting and enhanced light-shadow contrast textures. It requires the Majicflus model for optimal performance, and using it with Flux_dve may not yield satisfactory results.
I personally use Forge for image generation, and users can customize their workflow with Majicflus as needed.
Technical Parameters:
-
CFG: Default 3.5
-
Sampling steps: Default 20, can be increased to 30 if needed
-
LoRA weight: 0.7
-
Additional face and hand refinement (AD) is recommended for portraits with smaller face/body proportions
Trigger word: dingdaer
Example prompt: A girl in black clothing stands by a dark wall, with light and shadow falling on half of her face, creating a strong contrast between light and dark. Light shines through, and smoke fills the air.,dingdaer
"MAJICFLUS_麦橘丁达尔" Model (hereinafter referred to as "the Model") is developed and released by [DonRat] (hereinafter referred to as "the Owner"), based on MajicFlux's FLUX-lora model.
License Terms:
The Owner grants individuals and organizations permission to use images generated by this model free of charge for non-commercial educational or informational purposes, provided that:
-
Users comply with relevant laws and do not infringe upon the legal rights of the model or any third parties
-
Attribution must be given as "Generated by [DonRat]'s 'MAJICFLUS_麦橘丁达尔' Model" when using the images
For commercial use, a commercial license agreement must be signed with the Owner. For commercial licensing and custom model inquiries, please contact the Owner through their profile information.
The Owner will continue to provide free model updates to individual users as a gesture of support for community open-source contributors. Commercial user partnerships help drive the development and continuous improvement of this model. We appreciate every user's understanding and support.
训练词语: dingdaer
名称: 麦橘丁达尔.safetensors
大小 (KB): 149673
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success