metadata
tags:
- text-to-image
- lora
- diffusers
- template:diffusion-lora
- GArt
- Outdoor-Art
widget:
- text: >-
GArt, a man in a blue jacket and blue jeans is riding a brown horse down a
dirt road in the middle of a desert. The man is wearing a black cowboy
hat, a blue long sleeved shirt, and black boots. The horse is facing the
left side of the road, and the sun is setting in the background, casting a
shadow on the road. The road is divided into two sections, each section
having a yellow line painted on it. To the left and right of the image,
there are cactus plants with yellow flowers growing on the ground. In the
distance, there is a mountain range. The sky is a deep orange, with a few
wispy clouds in the sky.
output:
url: images/G1.png
- text: >-
GArt, Captured at night, a vibrant orange van is parked on a sandy beach.
The van is facing the viewer, with its door open, revealing the interior
of the van. A fire is burning on the beach, creating a warm glow. Two
folding chairs are positioned in front of the fire, adding a pop of color
to the scene. The sky is a deep blue, dotted with white stars, and a
mountain range can be seen in the background.
output:
url: images/G2.png
- text: >-
GArt, Captured from a high-angle perspective on a sunny day, a view of a
baseball stadium is seen. The stadium is surrounded by lush green grass
and trees, with a winding track running along the center of the field. The
field is adorned with a few people playing a game of tennis, and there are
a few empty stands on either side of the track. In the distance, a city
can be seen, with tall buildings and a mountain range in the background.
The sky is a deep blue, dotted with white clouds.
output:
url: images/G3.png
base_model: black-forest-labs/FLUX.1-dev
instance_prompt: GArt
license: creativeml-openrail-m
Flux-GArt-LoRA
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-GArt-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 & 2500 |
Epoch | 10 | Save Every N Epochs | 1 |
Labeling: florence2-en(natural language & English)
Total Images Used for Training : 14
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-GArt-LoRA"
trigger_word = "GArt"
pipe.load_lora_weights(lora_repo)
device = torch.device("cuda")
pipe.to(device)
Trigger words
You should use GArt
to trigger the image generation.
Download model
Weights for this model are available in Safetensors format.
Download them in the Files & versions tab.