Sign In

Oga_Ka

12

99

3

2

Verified:

SafeTensor

Type

LoRA

Stats

99

3

30

Reviews

Published

Oct 2, 2024

Base Model

Flux.1 D

Training

Steps: 250
Epochs: 5

Usage Tips

Clip Skip: 1
Strength: 1

Hash

AutoV2
D6C3E133A6

Created on Civitai

The FLUX.1 [dev] Model is licensed by Black Forest Labs. Inc. under the FLUX.1 [dev] Non-Commercial License. Copyright Black Forest Labs. Inc.

IN NO EVENT SHALL BLACK FOREST LABS, INC. BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH USE OF THIS MODEL.

Ogawa Kazumasa (1860-1929) was a Meiji era photographer, printer, and publisher, who was a pioneer in the field of photography and photomechanical printing.

Following a post on reddit (I know never trust what you read on reddit) this lora was a kind of experiment to see how well flux would train without any caption using only 10 images that I found under free license. Honestly I'm surprised, but it does add an interesting old printed art look.

Use the ogaka trigger if you like.