metadata
tags:
- text-to-image
- lora
- diffusers
- template:diffusion-lora
- flux-dev
- ultra
- realism
- photorealism
- hi-res
- face
- diffusion
- UltraRealism
widget:
- text: >-
Woman in a red jacket, snowy, in the style of hyper-realistic portraiture,
caninecore, mountainous vistas, timeless beauty, palewave, iconic,
distinctive noses --ar 72:101 --stylize 750 --v 6
output:
url: images/3.png
- text: >-
Photograph, candid shot, famous randomly couch and randomly finished with
randomly cats, center point for cat, Use camera is Canon EOS 5D Mark IV
with a Canon EF 24mm f/1.4L II USM lens, set at aperture f/2.8 for a depth
of field that highlights the furniture clean lines with rich and many
detail, randomly color and finished, soft ambient light, studio light
setting, ultra realistic, UHD, many details --chaos 1 --ar 9:16 --style
raw --stylize 750
output:
url: images/5.png
- text: >-
High-resolution photograph, woman, UHD, photorealistic, shot on a Sony
A7III --chaos 20 --ar 1:2 --style raw --stylize 250
output:
url: images/4.png
base_model: black-forest-labs/FLUX.1-dev
instance_prompt: Ultra realistic
license: creativeml-openrail-m
Canopus-LoRA-Flux-UltraRealism-2.0
- The new versions available: https://huggingface.co/prithivMLmods/Ton618-Epic-Realism-Flux-LoRA
- Also Available in Flux LoRA DLC π₯³: https://huggingface.co/spaces/prithivMLmods/FLUX-LoRA-DLC
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/Canopus-LoRA-Flux-UltraRealism-2.0
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 | 30 & 3.8K+ |
Epoch | 20 | Save Every N Epochs | 1 |
Labeling: florence2-en(natural language & English)
Total Images Used for Training : 70 [ Hi-RES ] & More ...............
Image Compare DLC.
Image | Description |
---|---|
Image 1 | Portrait of an attractive woman: Late twenties, light brown hair, wearing a yellow sweater, looking directly at the camera. Standing outdoors near trees. Aspect ratio: 128:85. Style: Raw |
Image 2 | Headshot of a handsome young man: Dark gray sweater with buttons and shawl collar, brown hair, short beard. Serious expression on a black background, soft studio lighting. Aspect ratio: 85:128. Style: Raw |
Image 3 | Fashion photo: Model in a white bodysuit and beige trench coat, posing with hands on head in front of a train station at sunset. High resolution, 35mm lens, f/22, natural lighting. Aspect ratio: 85:128. Style: Raw |
Image 4 | Rustic portrait: Woman with fair skin and natural, wavy hair in soft curls. Wearing a red and navy plaid shirt with a white undershirt, sleeves rolled up. Leaning against a weathered blue door frame with a contemplative expression. |
Other Sample
Trigger words
You should use Ultra realistic
to trigger the image generation.
Other Versions
Hereβs a table format for the Hugging Face model "prithivMLmods/Canopus-LoRA-Flux-FaceRealism":
Attribute | Details |
---|---|
Model Name | Canopus-LoRA-Flux-FaceRealism |
Model ID | prithivMLmods/Canopus-LoRA-Flux-FaceRealism |
Hugging Face URL | Canopus-LoRA-Flux-FaceRealism |
Model Type | LoRA (Low-Rank Adaptation) |
Primary Use Case | Face Realism image generation |
Supported Framework | Hugging Face Diffusers |
Data Type | bfloat16 , fp16 , float32 |
Compatible Models | Stable Diffusion, Flux models |
Model Author | prithivMLmods |
LoRA Technique | LoRA for image style transfer with a focus on generating realistic faces |
Model Version | Latest |
License | Open-Access |
Tags | LoRA, Face Realism, Flux, Image Generation |
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/Canopus-LoRA-Flux-UltraRealism-2.0"
trigger_word = "Ultra realistic"
pipe.load_lora_weights(lora_repo)
device = torch.device("cuda")
pipe.to(device)
App File Structure
/project-root/
β
βββ .gitattributes
βββ README.md
βββ app.py
βββ pythonproject.py
Download model
Weights for this model are available in Safetensors format.
Download them in the Files & versions tab . . .π€: https://hf.co/prithivmlmods