333-PRO

Prompt

333 Pro Sketch – A black and white drawing of a ram on a textured gray background. The ram’s head is turned to the left, showcasing its curved, pointed horns. Its face is detailed with fine lines, highlighting its white fur, black nose, and slightly open mouth. The ram’s eyes are black, giving it a strong, focused expression. Perched atop its head is a black bird, its feathers sketched with delicate strokes. The background remains minimal, allowing the intricate linework of the ram and bird to stand out, creating a striking contrast.

Prompt

333 Pro Sketch, A black and white drawing of a man playing a trumpet. The man is wearing a striped shirt and shorts. He is holding the trumpet in his right hand. His left hand is raised in the air. His right arm is extended out to the right. The background is a light beige color. The drawing is done in black ink.

Prompt

333 Pro Sketch, A detailed sketch of a man playing a saxophone on a light colored background. The man is wearing a short sleeve shirt and has short dark hair. His eyes are closed and his mouth is slightly open. The saxophone is held in his right hand and is positioned to the right of the saxophone. His left hand is positioned in front of his mouth. His right hand is resting on his hip. He has a beard and mustache. His hair is dark brown.

Model description for 333-PRO

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	25 & 2950
Epoch	23	Save Every N Epochs	1

Labeling: florence2-en(natural language & English)

Total Images Used for Training : 33

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 = "strangerzonehf/333-PRO"
trigger_word = "333 Pro Sketch"  
pipe.load_lora_weights(lora_repo)

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

Trigger words

You should use 333 Pro Sketch to trigger the image generation.

Download model

Weights for this model are available in Safetensors format.

Download them in the Files & versions tab.