Core ML Converted Model:

This model was converted to Core ML for use on Apple Silicon devices. Conversion instructions can be found here.
Provide the model to an app such as Mochi Diffusion Github / Discord to generate images.
split_einsum version is compatible with all compute unit options including Neural Engine.
original version is only compatible with CPU & GPU option.
Custom resolution versions are tagged accordingly.
The vae-ft-mse-840000-ema-pruned.ckpt VAE is embedded into the model.
This model was converted with a vae-encoder for use with image2image.
This model is fp16.
Descriptions are posted as-is from original model source.
Not all features and/or results may be available in CoreML format.
This model does not have the unet split into chunks.
This model does not include a safety checker (for NSFW content).

beautypromix-v1:

Source(s): CivitAI

關於BeautyProMix模型(人物及光影加強)

本模型經過約略1個月進行各項參數的調整並搭配大量的圖庫進行訓練，前後進行了近百次的混合及比例調整，此版本針對人像及光影表面十分令人讚賞(可參考示意圖的光影效果)。尤其是這個模型加入了BeautyPro首席穿搭顧問及彩妝師團隊的協助，針對用於訓練的每張照片進行細節的調整，使得面部表現更為優良。

感謝參與此模型製作的人員，包含製作者及提供照片作為模型訓練的朋友們，本模型搭配了多個主流模型進行訓練並條列如下，製作團隊已經盡力找到各個用於訓練模型的源頭，但難以確認每一個模型及每一張用於訓練的圖片是否存在版權疑慮，若我們的模型對您個人或公司的權益造成損害，請立即留言通知我們，製作團隊將立即將相關內容於模型中完全移除後，發布修正版本，謝謝。

#BeautyProMix *為了感謝無日無夜測試模型的製作者團隊及無私提供照片的所有朋友，請於製作出來的圖片標註是採用 #BeautyProMix，僅代表製作者團隊群感謝您。

About the BeautyProMix Model(Enhancement of Characters and Lighting)

After approximately one month of adjusting various parameters and training with a large database of images, this model has undergone nearly 100 blends and ratio adjustments. This version focuses on enhancing the appearance of characters and lighting, and has received high praise. In particular, the model has incorporated the expertise of the BeautyPro team's chief fashion consultants and makeup artists to fine-tune the details of each photo used for training, resulting in superior detail of the face.

We would like to thank everyone who contributed to the production of this model, including the creators and friends who provided photos for training. The model was trained with multiple mainstream models, and we have made every effort to locate the sources of each model used for training. However, it is difficult to confirm whether each model or image used for training may have copyright concerns. If our model causes any damage to your personal or company interests, please inform us immediately. The developing team will promptly remove the relevant content from the model and release a corrected version. Thank you.

#BeautyProMix

*To express gratitude to the develop team who tirelessly tested the model day and night, and to all friends who generously provided photos, please tag the images produced using #BeautyProMix. This tag represents the collective appreciation of the developing team.

Mixed-model list

BeautyProAsiaFace https://huggingface.co/beautypro/BeautyProAsiaFace

MUSE_v1 https://civitai.com/models/13564/musev1

AsiaFacemix https://huggingface.co/dcy/AsiaFacemix

Colorful https://civitai.com/models/7279/colorful

LOFI https://civitai.com/models/9052/lofi

Deliberate https://civitai.com/models/4823/deliberate

QGO-10b https://civitai.com/models/4188/qgo-10b

HassanBlend https://civitai.com/models/1173/hassanblend-1512-and-previous-versions

*重要備註：此模型完全未使用到明定無法商用的模型ChilloutMix及Dreamlike Diffusion進行訓練

*Important Note: This model was not trained using the ChilloutMix and Dreamlike Diffusion models, which are explicitly not available for commercial use.

徵求更多的好照片

製作團隊正在收集更多願意提供用於訓練的人物照片或景色圖片，若您願意提供給BeautyPro團隊進行訓練，團隊將在共同製作者名單中，加入您的名字，以展示您對於AI繪圖及相關領域的付出及努力！

We need more quality Photos

The developing team is currently collecting more character portraits and scenic images from those willing to provide them for training. If you are willing to offer your photos to the BeautyPro team for training purposes, your name will be included in the list of co-creators to showcase your contribution and efforts to the field of AI drawing.

即將推出

人物強化的進階版正在進行訓練中，若有更新的突破，我們將立即上傳提供大家測試及使用！若有任何建議及使用上的問題，也歡迎隨時留言，製作團隊感謝您提供的寶貴意見。

To be continued

An advanced version of the character and lighting enhancement model is currently being trained. If there are any breakthroughs in the update, we will immediately upload it for testing and use. If you have any suggestions or issues using the model, please feel free to leave a message. The developing team appreciates your valuable feedback.