Flux.1-Dev-Pov-DoorEye-LoRA

Prompt

look in 2, Captured at eye-level, a full view of two cartoon characters, a yellow cartoon character with big eyes and a wide mouth, stands on a gray carpeted floor. The character on the left is wearing a blue and white striped shirt, and black shoes. His arms are out to his sides, and he has a smile on his face. His head is turned to the right, and his eyes are open, as if he is looking at the camera. The creature on the right is facing the camera, and its mouth is open. They are standing in front of an elevator, which is gray in color. The elevator is surrounded by brown wooden walls, and there is a white ceiling above the elevator.

Prompt

look in 2, An eye-level view of a snoopy doll wearing a pair of black headphones. The doll is sitting on top of a piece of paper with a pencil in its right hand. There is a yellow remote control to the right of the doll. Behind the doll is a wooden table with a white and brown striped curtain on it.

Prompt

look in 2, Captured at eye-level, a close-up view of a blue cartoon robot with two black round eyes and a yellow plus sign on its chest. The robot is standing on a tan carpeted floor. The ceiling above the robot is illuminated by a bright white light. To the right of the robot, there is a red button with a yellow arrow pointing to the right.

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.1-Dev-Pov-DoorEye-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	15 & 2000
Epoch	12	Save Every N Epochs	1

Labeling: florence2-en(natural language & English)

Total Images Used for Training : 13

Best Dimensions

768 x 1024 (Best)
1024 x 1024 (Default)

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.1-Dev-Pov-DoorEye-LoRA"
trigger_word = "look in 2"  
pipe.load_lora_weights(lora_repo)

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