MajicFlus is a model fine-tuned and merged based on flux.dev, specializing in high-quality portrait generation, with a particular focus on capturing the delicacy and beauty of Asian female. The model emphasizes aesthetics, realism, and user-friendliness, allowing users to generate high-quality results with simple prompts while also handling complex prompts effectively.<h3 id="model-features-ydp4q1643">Model Features</h3><ul><li>Outstanding Portrait Generation: Optimized for various lighting conditions, ensuring facial details and limb integrity in different compositions.</li><li>Versatile Applications: In addition to portraits, the model demonstrates significant improvements in generating non-human subjects and scenes, making it suitable for diverse creative needs.</li><li>Ease of Use: Users can achieve impressive results with simple prompts, while longer prompts are also well-supported for fine-grained control.</li></ul><h3 id="community-adaptation-vvt0mudkt">Community Adaptation</h3>MajicFlus is launching alongside a collection of LoRAs created by multiple community members. These LoRAs further enhance the model’s functionality and performance, offering users more creative possibilities and enabling the model to adapt to specific styles and scenarios.<h3 id="weakness-71gtbu3nl">Weakness</h3><ul><li>MajicFlus is not designed for generating NSFW content. however, if needed, you can use anatomy LoRAs for such purposes.</li><li>MajicFlus exists to address the lack of Asian representation in models within the international community. If you wish to generate images of non-Asian ethnicities, please consider using other high-quality models.</li></ul><hr />麦橘超然 MajicFlus 是一款基于 flux.dev 微调融合的模型，专注于高质量人像生成，尤其擅长表现 亚洲女性 的细腻与美感。模型以 唯美、写实、易用 为核心特色，能够通过简单的提示词生成优质效果，同时对复杂提示词也有出色的响应能力。<h3 id="-3fydqu75i">模型特点</h3><ul><li>卓越的人像生成能力： 优化了在不同光影条件下的表现，确保人像在各种构图中的 面部细节 和 肢体完整性。</li><li>广泛的适用性： 除了人像生成外，模型在生成 非人生物 和 场景 时也有显著改进，适应更多创作需求。</li><li>简单易用： 用户无需复杂的提示词即可生成高质量作品，同时支持更长提示词的精细控制。</li></ul><h3 id="-w0g9buap7">社区适配</h3>MajicFlus 模型在发布的同时，多位社区成员基于模型制作的 LoRA 也将一同发布，进一步扩展了模型的功能与表现力。这些 LoRA 为用户提供了更多样化的创作可能性，使模型能够适应更多特定场景和风格需求。<h3 id="-4gj0tsb83">弱点</h3><ul><li>MajicFlus 并非为生成 NSFW 内容而设计。然而，如果有需要，可以使用相关 LoRA 来实现此类目的。</li><li>MajicFlus 的存在是为了解决国际社区中模型缺乏亚洲代表性的问题。如果您希望生成非亚洲种族的图像，请考虑使用其他高质量模型。</li></ul><hr /><h3 id="parameter-recommendation-mx5n896nc">参数推荐 Parameter Recommendation</h3><ul><li>Steps: 20~30</li><li>Distilled CFG Scale: 3.5</li><li>CFG : 1</li><li>Diffusion in Low Bits: float8-e4m3fn</li><li>Sampling: Euler + simple/beta (for general)，DPM2M + SGM uniform (for skin texture)，DEIS + DDIM uniform (for casual realistic look)</li><li>Vae: flux vae</li><li>Clip: clip_l.safetensors and t5xxl_fp8_e4m3fn.safetensors</li></ul>

majicFlus 麦橘超然

majicflus_v10.safetensors

v1.0

麦橘超然 MajicFlus 是一款基于&nbsp;<a target="_blank" rel="noopener noreferrer nofollow" href="http://flux.dev">flux.dev</a>&nbsp;微调融合的模型，专注于高质量人像生成，尤其擅长表现&nbsp;亚洲女性&nbsp;的细腻与美感。模型以&nbsp;唯美、写实、易用&nbsp;为核心特色，能够通过简单的提示词生成优质效果，同时对复杂提示词也有出色的响应能力。模型特点<ul><li>卓越的人像生成能力： 优化了在不同光影条件下的表现，确保人像在各种构图中的 面部细节 和 肢体完整性。</li><li>广泛的适用性： 除了人像生成外，模型在生成 非人生物 和 场景 时也有显著改进，适应更多创作需求。</li><li>简单易用： 用户无需复杂的提示词即可生成高质量作品，同时支持更长提示词的精细控制。</li></ul>社区适配与扩展MajicFlus 模型在发布的同时，多位社区成员基于模型制作的&nbsp;LoRA&nbsp;也将一同发布，进一步扩展了模型的功能与表现力。这些 LoRA 为用户提供了更多样化的创作可能性，使模型能够适应更多特定场景和风格需求。

麦橘超然 MajicFlus 

麦橘超然MajicFlus 是一款基于flux.dev 微调融合的模型，专注于高质量人像生成，尤其擅长表现亚洲女性的细腻与美感。模型以唯美、写实、易用为核心特色，能够通过简单的提示词生成优质效果，同时对复杂提示词也有出色的响应能力。
模型特点- 卓越的人像生成能力：
优化了在不同光影条件下的表现，确保人像在各种构图中的面部细节和肢体完整性。
- 广泛的适用性：
除了人像生成外，模型在生成非人生物和场景时也有显著改进，适应更多创作需求。
- 简单易用：
用户无需复杂的提示词即可生成高质量作品，同时支持更长提示词的精细控制。
社区适配与扩展MajicFlus 模型在发布的同时，多位社区成员基于模型制作的LoRA 也将一同发布，进一步扩展了模型的功能与表现力。这些LoRA 为用户提供了更多样化的创作可能性，使模型能够适应更多特定场景和风格需求。
- Steps: 20~30
- Distilled CFG Scale: 3.5
- CFG : 1
- Diffusion in Low Bits: float8-e4m3fn
- Sampling: Euler + simple/beta (for general)，DPM2M + SGM uniform (for skin texture)，DEIS + DDIM uniform (for casual realistic look)
- Vae: flux vae
- Clip: clip_l.safetensors and t5xxl_fp8_e4m3fn.safetensors

majicflus_v1.safetensors

麦橘超然majicFlus

MajicFlus is a model fine-tuned and fused based on flux.dev, focusing on high-quality portrait generation, especially excelling in portraying the delicacy and beauty of Asian women. The model is characterized by its aesthetic, realistic, and user-friendly features, capable of generating high-quality results with simple prompts while also demonstrating excellent responsiveness to complex prompts.<h3 style="text-align: left;">Model Features</h3><ul><li style="text-align: left;">Exceptional portrait generation capability: Optimized for performance under various lighting conditions, ensuring facial details and limb integrity in different compositions.</li><li style="text-align: left;">Wide applicability: Besides portrait generation, the model has shown significant improvements in generating non-human creatures and scenes, catering to more creative needs.</li><li style="text-align: left;">Simple and easy to use: Users can generate high-quality works without complex prompts, while also supporting fine control with longer prompts.</li></ul><h3 style="text-align: left;">Community Adaptation</h3>Upon the release of the MajicFlus model, several community members will also release LoRA models based on it, further expanding the model's functionality and expressiveness. These LoRA models provide users with more diverse creative possibilities, enabling the model to adapt to more specific scenarios and style requirements.<h1 style="text-align: left;">Recommended Parameters</h1><ul><li style="text-align: left;">Steps: 20~30</li><li style="text-align: left;">Distilled CFG Scale: 3.5</li><li style="text-align: left;">CFG: 1</li><li style="text-align: left;">Diffusion in Low Bits: float8-e4m3fn</li><li style="text-align: left;">Sampling: Euler + simple/beta (for general), DPM2M + SGM uniform (for skin texture), DEIS + DDIM uniform (for casual realistic look)</li><li style="text-align: left;">Vae: flux vae</li><li style="text-align: left;">Clip: clip_l.safetensors and t5xxl_fp8_e4m3fn.safetensors</li></ul>

Files with SHA256: 062dad72b2fee99af172463b3a2c84c1da793aaf80eaa0543138e05be55cdbbf