Downloads · 30 days
56
2% of all-time downloads
Castor-Character-Polygon-Flux-LoRA
prithivMLmods/Castor-Character-Polygon-Flux-LoRA
Castor-Character-Polygon-Flux-LoRA is a text-to-image model from prithivMLmods. Use it when you need an image from a text prompt. It is set up for diffusers. The card lists the license as creativeml-openrail-m.
Check the demo here: https://huggingface.co/spaces/prithivMLmods/FLUX-LoRA-DLC
All-time downloads
2.6K
Public
Repo size
615 MB
Likes
14
Public
Repo makeup
Click a slice to open those files.
.safetensors613 MB · 100%
At a glance
- Task
- Text-to-Image
- Library
- diffusers
- License
- creativeml-openrail-m
- Access
- Public
- Created
- Oct 23, 2024
- Updated
- Oct 23, 2024
- SHA
- b75fb2cc
Try a prompt
Base models
Model card
From the Hugging Face model README
Castor-Character-Polygon-LoRA-Flux
Check the demo here: https://huggingface.co/spaces/prithivMLmods/FLUX-LoRA-DLC
<Gallery />The model is still in the training phase. This is not the final version and may contain artifacts and perform poorly in some cases.
Model description
prithivMLmods/Castor-Character-Polygon-LoRA
Image Processing Parameters
| Parameter | Value | Parameter | Value |
|---|---|---|---|
| LR Scheduler | constant | Noise Offset | 0.03 |
| Optimizer | AdamW | Multires Noise Discount | 0.1 |
| Network Dim | 64 | Multires Noise Iterations | 10 |
| Network Alpha | 32 | Repeat & Steps | 27 & 1.9K+ |
| Epoch | 15 | Save Every N Epochs | 1 |
Labeling: florence2-en(natural language & English)
Total Images Used for Training : 25+ [ Hi-RES ]
Data Source
https://search.muz.li/
Trigger words
You should use 3D Polygon to trigger the image generation.
You should use 3D Polygon Character to trigger the image generation.
Character Creation Prompts
| Character Type | Description |
|---|---|
| Hyper-Realistic Young Man | "Create a hyper-realistic polygon 3D portrait of a young man in his late 20s. He has short, dark hair, deep blue eyes, and light stubble on his face. He wears a red cap, slightly worn, with a casual, loose-fitting t-shirt. His expression is neutral, with a focused gaze. The lighting highlights his detailed facial textures, showing light freckles and a slight sheen on his skin, adding to the realism. The background is out of focus, with industrial elements, giving a gritty, urban feel." |
| Elegant Woman in a Red Dress | "Create a hyper-realistic polygon 3D portrait of an elegant woman in her early 30s. She has long, wavy blonde hair and wears a vibrant red evening gown with delicate earrings. Her expression is poised and confident, with soft makeup accentuating her features. The lighting highlights her clear skin and the slight shimmer of her dress. The background is a soft-focus dark texture, drawing attention to her serene yet powerful presence." |
| Realistic Woman with Wet Hair | "Create a hyper-realistic polygon 3D portrait of a woman in her late 20s with dark brown, wet hair slicked back. She has golden-brown eyes and subtle freckles on her face. She wears minimal makeup, and her skin shows natural imperfections, adding to the realism. Her expression is neutral yet focused, with slightly parted lips. The lighting is soft and even, emphasizing the texture of her skin and wet hair. The background is a simple gray gradient, keeping the focus on her." |
| Modern-Day Woman with Short Hair | "Create a hyper-realistic polygon 3D portrait of a woman in her early 30s with short, tousled platinum blonde hair. She has piercing blue eyes, pale skin with subtle freckles, and a sharp jawline. She wears a black leather jacket over a simple white t-shirt, giving off a contemporary, urban vibe. Her expression is confident and slightly rebellious, with a smirk playing on her lips. The lighting is dramatic, casting shadows on one side of her face and highlighting the texture of her skin and the gloss on her lips. The background is a cityscape at night with blurred lights, adding to the modern and edgy atmosphere." |
| Stylish Man with Beard and Glasses | "Create a hyper-realistic polygon 3D portrait of a man in his mid-30s with neatly trimmed brown hair and a full beard. He wears black-rimmed glasses and a navy-blue suit with a crisp white shirt. His expression is serious yet approachable, with a slight furrow in his brow. The lighting highlights the texture of his beard and the slight glare on his glasses. The background is a blurred modern office environment, giving a professional yet relaxed tone." |
Sample Image + Prompt
| Prompt |
|---|
| Create multiple side-angle shots of a highly realistic 3D polygon character portrait featuring a young woman with vibrant curly red hair and a light complexion. Her hair is voluminous, showcasing a variety of textured curls that catch the light beautifully. She has a face adorned with delicate freckles scattered across her cheeks and nose, contributing to her youthful appearance. Her striking green eyes are large and expressive, framed by long lashes that add depth to her gaze. She possesses soft, naturally colored lips that are slightly parted, enhancing her inviting demeanor. The character has an oval face shape with soft contours, complemented by a petite nose that adds to her delicate features. The side views should highlight the luxurious volume of her curls, allowing for dynamic play of light and shadow, emphasizing the texture and richness of her hair. The background should be neutral and unobtrusive, designed to enhance the vibrant red hues in her hair and the softness of her complexion. Soft, diffused lighting should be used to create an ethereal glow on her skin, emphasizing her freckles and adding a warm undertone that complements her overall look. Focus on capturing the intricate details of her features, including the reflections in her green eyes, the subtle blush on her cheeks, and the gentle curvature of her lips. |
Setting Up
import torch
from pipelines import DiffusionPipeline
base_model = "black-forest-labs/FLUX.1-dev"
pipe = DiffusionPipeline.from_pretrained(base_model, torch_dtype=torch.bfloat16)
lora_repo = "prithivMLmods/Castor-Character-Polygon-LoRA"
trigger_word = "3D Polygon" # Leave trigger_word blank if not used.
pipe.load_lora_weights(lora_repo)
device = torch.device("cuda")
pipe.to(device)
App File Structure
/project-root/
├── .gitattributes
├── README.md
├── app.py
├── pythonproject.py
Download model
Weights for this model are available in Safetensors format.
Download them in the Files & versions tab.
- Task
- Text-to-Image
- Library
- diffusers
- License
- creativeml-openrail-m
- Fal Ai
- live
- Replicate
- live
- Wavespeed
- live
- Created
- Oct 23, 2024
- Updated
- Oct 23, 2024