Flux-C33-Design-LoRA

Prompt

C33 Design, a close-up shot captures a vibrant blue smiley face, adorned with a black hat adorned with two yellow buttons. The hat is adorned with an eye-level yellow circle with a smiling face, adding a pop of color to the scene. The eyes are glowing blue, creating a striking contrast to the black hat. The helmet is adorned by a black strap, adding depth to the composition. The background is a muted gray, adding to the overall composition.

Prompt

C33 Design, A 3D rendering of a persons head and neck is visible. The person is wearing a white long sleeved t-shirt with a blue stripe at the bottom, and a black glove on their left hand. They are holding a white stick in their right hand. The background is a vibrant red, and there is a circular object in the center of the persons neck that has three black circles on it. There is a gold ring on the top left corner of the image.

Prompt

C33 Design, a vibrant purple smiley face is adorned with a small red tag that reads "Hello, I am a JPEG". The tag is affixed to the face of the purple face, adding a pop of color to the otherwise monochromatic figure. The figure is wearing a dark blue jacket, with a black backpack strapped to its back. The background is a stark white, creating a stark contrast to the figure.

Prompt

C33 Design, a mannequin is adorned with a white hat adorned with black dots and a smiley face. The hat has a yellow circle button in the center of it, adding a pop of color to the scene. The backdrop is a vibrant orange, creating a striking contrast to the white hat and the black jacket.

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/Flux-C33-Design-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	18 & 2450
Epoch	14	Save Every N Epochs	1

Labeling: florence2-en(natural language & English)

Total Images Used for Training : 17

Best Dimensions & Inference

Dimensions	Aspect Ratio	Recommendation
1280 x 832	3:2	Best
1024 x 1024	1:1	Default

Inference Range

• Recommended Inference Steps: 30–35

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/Flux-C33-Design-LoRA"
trigger_word = "C33 Design"  
pipe.load_lora_weights(lora_repo)

device = torch.device("cuda")
pipe.to(device)

Trigger words

You should use C33 Design to trigger the image generation.

Download model

Weights for this model are available in Safetensors format.

Download them in the Files & versions tab.