metadata
license: other
license_name: flux-1-dev-non-commercial-license
license_link: https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md
language:
- en
tags:
- flux
- diffusers
- lora
- replicate
base_model: black-forest-labs/FLUX.1-dev
pipeline_tag: text-to-image
instance_prompt: KUJI
widget:
- text: Cherry tree in the style of KUJI
output:
url: examples/1.JPG
- text: Vibrant sunset. Effects and flares in the style of KUJI.
output:
url: examples/2.JPEG
- text: >-
View through car window looking out on the desert, light leak effect,
scenery, blurry, flmft in the style of KUJI.
output:
url: examples/3.JPG
- text: >-
A mountain peak piercing through a sea of clouds at sunrise, with
alpenglow on the summits, lens flare streaking across, flmft in the style
of KUJI
output:
url: examples/4.JPG
- text: Sunny day at the beach in the style of KUJI.
output:
url: examples/5.WEBP
- text: >-
Teenage bedroom with posters and clutter, warm color palette, flmft in the
style of KUJI.
output:
url: examples/6.JPG
- text: Mount fuji landscape in the style of KUJI.
output:
url: examples/7.JPG
- text: >-
A pier extending into a calm ocean, reflected in the water, with a sky
full of cotton candy cirrocumulus clouds at sunset, dreamy double exposure
effect, flmft in the style of KUJI
output:
url: examples/8.JPG
- text: >-
A lone tree on a hilltop, framed by a sky with layered lenticular clouds
catching the last light of day, vignette effect, flmft in the style of
KUJI
output:
url: examples/9.JPG
Flux Lora Kuji
A Huji cam inspired lora finetuning of flux-koda. Works best with landscapes and naturalistic environments. In order to obtain better results, combine the keywords 'flmft' and 'in the style of KUJI'. Fun fact: training set was mostly made by images take by me 🤗.
Trained on Replicate using:
https://replicate.com/ostris/flux-dev-lora-trainer/train
Trigger words
Combine the keywords 'flmft' and 'in the style of KUJI'
Use it with the 🧨 diffusers library
from diffusers import AutoPipelineForText2Image
import torch
pipeline = AutoPipelineForText2Image.from_pretrained('black-forest-labs/FLUX.1-dev', torch_dtype=torch.float16).to('cuda')
pipeline.load_lora_weights('ludocomito/flux-lora-kuji', weight_name='lora.safetensors')
image = pipeline('your prompt').images[0]
For more details, including weighting, merging and fusing LoRAs, check the documentation on loading LoRAs in diffusers