Sign In

Hyper Detailed Illustration

292

2.1k

1.4k

147

Updated: Sep 22, 2024

stylerealismillustration

Verified:

SafeTensor

Type

LoRA

Stats

2,061

1.4k

7k

Reviews

Published

Sep 12, 2024

Base Model

Flux.1 D

Training

Steps: 800
Epochs: 8

Trigger Words

hyp3rd3tail style

Hash

AutoV2
7BCA99EA21
default creator card background decoration
toasters's Avatar

toasters

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.

This LoRA is designed to create a very detailed semi realistic illustration effect. Adjusting the weight can be really useful for changing the amount of the effect. Works very well with some prompts but not so well with others but usually adjusting the weight can help. Trigger is hyp3rd3tail style. Can have slightly different effects depending on the type of image you prompt for. I really recommend trying different weights especially if you get strange results.

Also it does seem to work without the trigger and at low weights can improve details of realistic images.

Please share your creations! Would love to see them and might help me decide on the direction to go with version 2.

And don't ask about the random cat ears.... I have no idea. Even my pumpkins wanted cat ears. But in those cases usually turn down the weight.

Having spoken to a few people using Forge and if having issues with LoRAs and using the NF4 model changing the Diffusion in Low Bits to "bnb-nf4 (fp16 LoRA)" instead of "Automatic" will probably solve the problem.

If you would like to buy me a coffee: https://ko-fi.com/kleebztech

Trained using Flux Gym which you can learn about in this video: