TikTok style lora

First, look at the instructions

This is the Lora version of the Douyin-style large model, still available for download for non-members.

This is a Lora model in a short video style, featuring beautiful women and handsome men selected by intelligent algorithms from short video platforms that are more suitable for the aesthetic preferences of Chinese people. The images produced by this model are also more suitable for creating videos, aligning with the aesthetic standards of short videos, don't you think?

This version focuses on enhancing learning related to facial features, body types, skin tones, and Douyin style.

You can trigger the model with or without specific keywords, including "dygirl".

This model has been refined by selecting nearly 500 popular short videos from the past two years (results may vary), and finally reducing them to 100 videos for frame sampling. Most extractions were done at 3-5 frames per second, removing ghosting and other visual disturbances. Each image was then manually edited (as dynamic frame shots are usually blurry) to create a training set (some images of the same person were partially removed to ensure training effectiveness).

This version has essentially addressed the issue of degraded hand features in distilled models. However, since the training set consists of sampled frames with manual edits, some flaws may still exist.

The example images are not part of the training set and are reverse-tagged. The prompt words are within the node "Display Text". Click on an example image, check the workflow, and locate the Display Text node.

This version is recommended for use with comfyui for optimal results.

Flock:

Future letter: 463664827

Model Usage Statement

1. You may not host this model and its derivative versions (such as fusion model versions) on websites or applications intended to generate income or donations. If you need to use this model and its derivative versions for commercial purposes (such as generative services, selling images, publicly posting images on short video platforms, articles or publications, etc.), please contact me via the site.

2. You may not directly sell images generated by this model unless you have made enough manual modifications to the image to be wholly considered your personal work in a legal sense. If you violate this clause, you are personally responsible for all legal consequences, and I will not be held liable.

3. You must not deliberately use this model to create or share illegal or harmful content for dissemination and output. Please abide by public morality and use this model for positive and proactive purposes.

4. To avoid misunderstandings, please do not upload this model to any other model website, do not use images produced by this model to train new models, and prohibit any form of fusion models. Thank you!