Sign In

Dark Infinity's Misc Flux LoRAs

92

Updated: Nov 23, 2024

style

Download

1 variant available

SafeTensor

292.24 MB

Verified:

Type
LoRA
Stats

341

82

516

Reviews
Published

Nov 23, 2024

Base Model

Flux.1 D

Training
Epochs: 13
Usage Tips
Strength: 1
Hash
AutoV2
10E5CB891A
Trigger Words
ditttdlgnc
default creator card background decoration
Reactions - 74750

74.8K

Followers - 2955

3K

Downloads - 90088

90.1K

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.

00168-3401782280.png

This model card is reserved for LoRAs that didn't quite cut it to be released on their own but ones I may want to use on the generator or have others download and test. I am putting it up front that these are random, weird, and sometimes broken models—if you download any to try, feedback is welcome, but results are not guaranteed.

Painterly-Realism v2.0

So, fun fact: painterly style and realism are hard to combine. Just because you like two things doesn't mean you should try to shove them together into a single style. I ended up with something that occasionally worked to push realism into digital art but lost the painterly. After multiple training attempts, it either overfitted or lost the painterly style. This was the best of the failures, but it still doesn't do much; it just makes Flux look a bit like a realist painting.

Evo-Expressions v3.1

I found a few paintings that inspired me to do an evolved take on expressionist art. Unfortunately, my dataset was extremely limited (4 images with very little variation) and though it turned out very well during training, I quickly discovered its ability to generalize is minimal. Its not a complete failure because it does what it does really well. It is great if you want to make modern, expressionist paint-splattered images of women, but pretty limited if you want to do anything else. It does do some nice animals though.

Fractal Flux v1.0

Trained on a huge fractal image, this LoRA can be fantastic, cause the image to look pixelated, or do nothing. It was an experiment using a 1-image LoRA with no captions.

Ethereal Rainbow v1.0

Trained in psychedelic art, this LoRA was poorly captioned (because how do you caption that?) but can make some exciting and unusual images. Does it do something? Yep, it causes a radical change for most images, putting color and shapes everywhere. Is it useful? Not really.