MYHuman-墨幽随拍-Flux-trainable
MYH-1.1
MoYou
New version of Xiaohongshu Tu Bang model
This is a daily shooting style model that highlights the realism of high p and high blur mesh images.
This version is a pure trainable version of the model that has not been integrated since training. It can be used as a base model training lora and can continue to be trained.
This version requires self loading of clip, t5xxl, and vae during runtime
webui-forge
(Sampler)：euler
(Schedule type)：Simple
(Steps)：20
Distilled CFG Scale：3.5
(CFG Scale)：1
Comfyui can directly use the default Flux workflow
QQ group:
Moyou Alchemy Pavilion - Taiqing Pavilion: 858495398
Moyou Alchemy Pavilion - Shangqing Pavilion: 764481772
Please also pay more attention to my Little Red Book and Tiktok account: yy469370733
Model Usage Declaration
1. You are not allowed to host this model and its derivative versions (such as fusion model versions) on websites/applications that plan to earn income or donate. If you need to use this model and its derivative versions for commercial purposes (such as generative services, selling images, using images for publishing articles or publications on short video platforms, etc.), please contact me through the website
2. You are not allowed to directly sell the images generated by this model unless you have manually modified the images to a sufficient extent that they can be fully recognized as your personal work in legal terms. If you violate this clause, all legal consequences caused shall be borne by you personally, and I am not responsible.
3. You cannot intentionally create or share illegal or harmful content dissemination and output using this model. Please abide by public order and good morals, and use this model for positive purposes.
What is MYHuman-墨幽随拍-Flux-trainable?
MYHuman-墨幽随拍-Flux-trainable is a highly specialized Image generation AI Model of type Safetensors / Checkpoint AI Model created by AI community user MoYou. Derived from the powerful Stable Diffusion (Flux.1 D) model, MYHuman-墨幽随拍-Flux-trainable has undergone an extensive fine-tuning process, leveraging the power of a dataset consisting of images generated by other AI models or user-contributed data. This fine-tuning process ensures that MYHuman-墨幽随拍-Flux-trainable is capable of generating images that are highly relevant to the specific use-cases it was designed for, such as photorealistic, sexy, base model.
With a rating of 0 and over 0 ratings, MYHuman-墨幽随拍-Flux-trainable is a popular choice among users for generating high-quality images from text prompts.
Can I download MYHuman-墨幽随拍-Flux-trainable?
Yes! You can download the latest version of MYHuman-墨幽随拍-Flux-trainable from here.
How to use MYHuman-墨幽随拍-Flux-trainable?
To use MYHuman-墨幽随拍-Flux-trainable, download the model checkpoint file and set up an UI for running Stable Diffusion models (for example, AUTOMATIC1111). Then, provide the model with a detailed text prompt to generate an image. Experiment with different prompts and settings to achieve the desired results. If this sounds a bit complicated, check out our initial guide to Stable Diffusion – it might be of help. And if you really want to dive deep into AI image generation and understand how set up AUTOMATIC1111 to use Safetensors / Checkpoint AI Models like MYHuman-墨幽随拍-Flux-trainable, check out our crash course in AI image generation.
Popularity
Info
Version MYH-1.1: 1 File
About this version: MYH-1.1
First, let's take a look at the instructions (click on 'Expand All' to expand all the instructions)
1.1：
This version is a dreambooth full fine-tuning model, which is a biased experimental model and has not shown any degradation performance in testing. The hand performance should be basically the same as the original dev.
The difference from 1.0: 1. Hands are better than v1, and when used alone, it should not be as user-friendly as v1 overall.
Can be paired with the MYH-1.1 version of Moyou-F.1-Lora-NetMap to achieve optimized hand capture effects.
Lazy people can directly use the v1 instant version.
2. V1 is a self training lora merge dev, resulting in severe hand degradation and compatibility degradation with lora. Due to different keys, it is not possible to use Akiba and Azure Dragon scripts as base models to train lora, and db as base models to continue training (this is a common drawback of all lora merge models that use common fusion methods).
1.1 is a dreambooth fully fine tuned model with dev as the base model, and no degradation has been found so far. It can be further trained or used as a base model to train lora. It also shows good compatibility with flux dev trained lora (realistic classes may require appropriate weight reduction), and the lora effect trained with itself as the base model is also comparable to that trained with dev as the base model.
3. Compatible with hyper-8 step acceleration model (click to download)
Quick to get started: (Using fp8 mode locally, the method is the same as using video memory and dev-fp8)
Hand gesture: This is a real high-definition photo, (add other after hand gesture)
Can be paired with the MYH-1.1 version of Moyou-F.1-Lora-NetMap to achieve optimized hand capture effects.
Parameters: Both online and local (webui forge, comfyui) are default.
webui-forge：
Sampling method (Sampler): Euler or Flux Realistic
Schedule type: Simple
Iteration Steps: 20
Distilled CFG Scale：3.5
Guidance coefficient for prompt words (CFG Scale): 1
Goose flock:
Moyou Alchemy Pavilion - Taiqing Pavilion: 858495398
Moyou Alchemy Pavilion - Shangqing Pavilion: 764481772
Please also pay more attention to my Little Red Book and Tiktok account: yy469370733
5 Versions
Go ahead and upload yours!
