For Flux1 Schnell users: you do not need the Lora, just the .txt training data
click here if you are looking for SDXL (Beta3 version)
BASE MODEL: Johnson Diffusion Zenkai
Zenkai System Explained: https://civarchive.com/articles/199
Update 24th May '23 - Added "DesuZenkaiV21beta2-lycoris" (beta2)
Update 21st May '23 - Added "djzJohnsonDesuZenkaiV15-lycoris" (beta1)
Animate Diff V2 video:
all new version built on the Johnson Diffusion "Zenkai" Dataset
Note:
Beta1 contained 230 style models
Beta2 contains 350 style models
Full Instructions, Prompt Examples, Caption lists and Wildcard Files added to Training Data section for both CKPT and LORA version.
The Zenkai System allow for easy interpolation of 230 of my original styles, multiplied by 5000 original prompts & 4200 generated variation prompts which can be mixed, doubled or tripled and used with txt2img or img2img with ease.
Art by Pure chance or by exact design. Your Choice.
-------------------------------------------
ALL BELOW IS OLD 108 LORA VERSION
-------------------------------------------
Update: 24th Feb '23 - Added "djzJohnsonDesuV21-320"
same as below, version 1, built on SD 2.1-768 with LORA 320.
Update: 23rd Feb '23 - Added "djzJohnsonDesuV15-320"
same as below, version 1, built on SD 1.5 with LORA 320.
Update: 22nd Feb '23 - Added "djzJohnsonDesuV21-256"
same as below, version 1, built on SD 2.1-768 with LORA 256.
Update: 18th Feb '23 - Added "djzJohnsonDesuV15"
same as below, version 1, built on SD1.5 with LORA 128.
This is "djzJohnsonDesuV2"
The result of retraining all of the djz Diffusion collection into a single LORA 128. While this contains the original 108 tokens, you do not need to use them and can prompt normally. While this is 250gb of models in a 200mb filesize, this will produce different images than other LORA or the original models. be sure to scroll down and get the links for the negative embeddings used to create the demo pics ;)
Version 1:
108 Concepts, 4277 images.
Epoch 1 - 42,770 Steps
Epoch 2 - 85,540 Steps
Links to recommended LORA/Negative Embeds below
Epoch 1 & Epoch 2 both gave nice and unique images, so we are releasing both versions.
LORA 128 Demo Epoch 1 vs Epoch 2 vs Base only (same prompt/seed)
Resources we use with our SD2.1-768 models:
nfixer, nartfixer, nfixernext - from Illuminati Diffusion
https://huggingface.co/IlluminatiAI/Illuminati_Diffusion_v1.0/tree/main/embeds/negative
Neg_Facelift768 by SoCalGuitarist
https://civarchive.com/models/2385/socalguitarists-magic-facelift-negative-embedding-for-model-2x-fix-yo-ugly-faces
DrD_PNTE768 by Dr.Diffusion
https://civarchive.com/models/4044/doctor-diffusions-point-e-negative-embedding
DangerGoose by Drift Johnson
https://civarchive.com/models/8107/djz-johnsons-image-helper-dangerhawk-and-dangergoose
Contrast Fix & High Key by theOvercomer8 (style LORA)
https://civarchive.com/models/8765/theovercomer8s-contrast-fix-sd21-768 https://civarchive.com/models/9205/to8s-high-key-lora-sd21-768
Resources we use with our SD1.5 models:
DangerHawk by Drift Johnson
https://civarchive.com/models/8107/djz-johnsons-image-helper-dangerhawk-and-dangergoose
Description
108 separate datasets were used to train this LORA 128 fine tune.
Recommend using with Negative Embeddings.
You do not need to use a Trigger Word.
Links in the main description!
FAQ
Comments (13)
Glad you listened to me at last
its not an extraction and big plans take time ;) this was always on our roadmap. thanks for your patience
@driftjohnson Being said that I have been testing extracting with DIM 8 and Network 8 and 1 and in some overtrained models the results were even better. Don't know exactly the tech side of why is causing that but I suggest to grab a hard overtrained model and make an extration not of 300 (which is also valid, see kappa_neuro), more like 8-8, 8-1 or 32-8 I can say that some times I got surprised. Of course I'm talking about SD 1.5 since didnt touch 2.1.
In the case you want to try, use Kohya_SS Lora extractor ;)
@davizca we have used the extractor, and all the models have already been converted and evaluated. However the demo images take time to complete. Kappa is also retraining these djz models properly from the original datasets. Individual models are easier to mix, so this is the general purpose model LORA version.
@axsthxticroot Oh are the Checkpoints "out of date now"? Ie is it better to just use these Loras?
I keep getting "Failure to match keys when loading Lora" error using epoc 1 and 2.
Are you sure you're using a v1.5 model with the v15 lora and not intermixing it with a 2.0 or 2.1 model/lora?
@Trahloc thats probably the problem? im not using anything 2.0
@Adellanuki yeah, grab the right versions to go with your base model
@scruffynerf lul, that's myb, thanks
Now this is something really interesting to test...
What are 108 concepts ? Please list them. And preview images for them, thanks.
if you type "djz" here on civitAI, you can find the full list by filtering for 2.1-768 Checkpoint models. Unfortunately there is not enough room in the Website UI to display them, so you can find them there.