Umamusume All in One LoRA
This is an All-in-One LoRA designed to generate a wide variety of characters from the popular game Umamusume: Pretty Derby with a single model. Without the hassle of managing individual character LoRAs, you can easily bring numerous Umamusume to life, including Stay Gold, Almond Eye, Loves Only You, and many more, with just this one file.
Notice
This model is a re-upload. The previous model page was accidentally deleted during a version update process. My apologies for the inconvenience!
✨ Features
Extensive Character Library: Capable of generating most of the main characters from the game with high fidelity.
Versatile Costume Support: Faithfully reproduces each character's unique racing outfits, as well as school uniforms, casual wear, and other styles.
Easy to Use: Achieve your desired images simply by inputting character names or feature tags, without the need for complex prompts.
🌟 Character & Prompt Information
For detailed prompt information regarding the characters, costumes, and features included in this LoRA, please refer to the Hugging Face link. Find the precise tags for the character you wish to create and utilize them in your prompts.
License
This model is released under "Derivative work guidelines for umamusume".
And "Fair-AI public license 1.0-SD" License
Description
This is alpha version of SDXL AIO LoRA of Uma musume
Recommended options
LoRA weight 1.0
Trigger words
[character name] \(umamusume\) or [character name]
For example,
manhattan cafe \(umamusume\)
hayakawa tazuna
SDXL version
trained on sd-scripts by kohya_ss and LyCORIS by KohakuBlueleaf
Thank you all a lot!
Base model : SDXL 1.0
Hardware : 4x RTX TITAN
Training time : 37 hours
Dataset : ULTIMA Dataset
Resolution : 768x1024 with aspect ratio bucketing
Dim, Alpha : 32
with dynamic pruning rank=24
Network dropout = 0.1
Optimizer : AdamW, weight decay=0.1
Steps : 12,024
Total Batch size : 48
4 GPU x 3 batch size x 4 Gradient Accumulation steps
Learning rate : warmup to 1e-4 for 340 steps and then kept constantLR scheduler
LR scheduler is constant_with_warmup.
Settings
For Stable Diffusion XL
CLIP skip = 1 or 2. Use whatever you want.
Use Hires.fix to get higher quality image
Upscaler
Latent, Latent (nearest-exact), Latent (bicubic antialiased) or other Latent series.
Denoising strength 0.50~0.65
If you want to use SDXL-alpha LoRA with AnimagineXL, use weight around 0.6
FAQ
Comments (19)
I need to thanks you for your hard works on sdxl version and thanks you to choose my algorithm and the base model!!!
Want to know if there are some chances for me to contact you directly?
Since we (me and cyber-meow a.k.a. alea31415) also love Umamusume and want to train more good model on it. If you need, I can also provide some computing resources and technical support.
If you think that's good please let me know!
Kohaku-BlueLeaf
😎
Can you update some chara outfit? like Silence Suzuka swimsuit
New costumes are in our plan. But it will take times. I think, at the end of L'Arc scenario, they will be updated
Katsuragi ace's prompt is weird!
The feature tags is weird.
Trigger word webpage!
ULTIMA-prompts check plz :)
I love you.
Is it true that yakuza death squads will come after you for using this lora?
🤣
HuggingFace's prompt,md has been deleted, where can I find it?
Any plans for PonyXL version?
Currently, I work on upgrading ULTIMA dataset. And some test on synthetic captioning using VLM such as LLaVA, gemini-vision, claude-3 family or GPT4V. But, I will try to train ponyXL based using peft methods(LoRA family). I think if I will train pony-XL based model, it will be an unstable beta version.
Nevertheless, I will try it, but what I want to focus from now, rebuilding or enhancing ULTIMA dataset for SD-3
@mht Thank you for your consideration!
PonyXL version is just released!
Will there be a PonyXL version released?
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.