metadata
tags:
- text-to-image
- lora
- diffusers
- template:diffusion-lora
- Knitted Character
- Thread
widget:
- text: >-
Knitted Character, A close-up view of a crocheted toy. The toy is made up
of a blend of orange, yellow, and pink yarn. It has a black button in the
center of its head. The background is a solid white color.
output:
url: images/KC1.png
- text: >-
Knitted Character, a crocheted doll stands on a wooden platform. The dolls
head is encased in a multi-colored knit outfit, adorned with a pair of
dark-colored goggles. The eyes are tinted blue, adding a pop of color to
the scene. The figures arms and legs are positioned in a way that creates
a sense of balance. The background is a stark white, creating a stark
contrast to the figure.
output:
url: images/KC2.png
- text: >-
Knitted Character, A close-up view of a knitted cat toy. The toy is a gray
and white striped cat with blue eyes. The cats nose is pink. Its eyes are
a piercing blue color. Its ears are made of gray yarn. Its whiskers are
black and white. Its paws are knitted with white yarn. The background is a
solid white color.
output:
url: images/KC3.png
base_model: black-forest-labs/FLUX.1-dev
instance_prompt: Knitted Character
license: apache-2.0
Knitted-Character-Flux-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/Knitted-Character-Flux-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 | 20 & 2000 |
Epoch | 10 | Save Every N Epochs | 1 |
Labeling: florence2-en(natural language & English)
Total Images Used for Training : 30 [ 4K Hi-RES ]
Best Dimensions
- 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/Knitted-Character-Flux-LoRA"
trigger_word = "Knitted Character"
pipe.load_lora_weights(lora_repo)
device = torch.device("cuda")
pipe.to(device)
Trigger words
You should use Knitted Character
to trigger the image generation.
Download model
Weights for this model are available in Safetensors format.
Download them in the Files & versions tab.