GREAT SCOTT! Vivid insanity is here!
===========================================================================
KR34M - 11/25 - Let's give FLUX/KREA the DR34M makeover. This checkpoint excels at photorealism and abstract artwork. We hope you enjoy what's likely our last FLUX release!
Note: NF4 is actually K4_S in GGUF format. FP8 is actually Q8_0, also GGUF!
We recommend the BF16 variant + T5 BF16 if you have the VRAM.
Note: Use LoRA with C4PACITOR at slightly lower than normal weights for best results and to avoid anatomy blending.
===========================================================================
C4PACITOR models are created by DR34MSC4PE with all the trappings you've come to expect as well as some cool bonuses:
Enhanced realism and photo-realistic concepts, trained on high quality datasets with the latest techniques.
Specific anime/illustration tuning to introduce some lost artistic concepts
NSFW tuned and capable of realism and artistic/anime images. Female anatomy is well represented with additional fine tuning in the works.
Exceptional performance with character/other/stacking Lora
Like our work? Buy us a coffee: https://ko-fi.com/dr34msc4pe
===========================================================================
DR34MSC4PE is
c0ur4ge (training/qa/inference code) /
eraser851 (training/captions/tooling code/data/qa)
===========================================================================
Recommendations by base model:
dev - (d_v1/d_v2/a_v)
===========================================================================
CFG: 3-7
Steps: 22-60
Sampler: RES_2M
Scheduler: Beta/Beta57/Bong Tangent
========================================
Hyper8/L16HT - (x-series)
===========================================================================
CFG: 3-6
Steps: 16-32
Sampler: DEIS
========================================
schnell - (s_v0/s_v1):
===========================================================================
Note: 100% Schnell Base
Steps: 2-14
Sampler: DEIS
===========================================================================
Description
Adjusted balance of concept weights based on feedback - should offer more balanced performance without sacrificing the ability of NSFW concepts using higher than standard Guidance Scale ( => 4.0 ).
FAQ
Comments (7)
will you be making a s v1.1 or v2?
Yup! Just need some time to generate the checkpoints and some images for the showcase.
I also need to make quants at some point in the coming days! Stay tuned!
@c0ur4ge thank you for the response!
s_v1.1 is up for your enjoyment! :)
@c0ur4ge thank you, its a fun model, ill be honest i hope not insulted when i say its terrible for realism quality, and also it tilts a bit too much in nsfw for realism (which has its uses) but i do like your models for some purposely absurd things.
@joehorse Not at all! I think this is probably a “branding” problem I should correct. I really don’t mean “photo realism” as I think is the case with many models on the site, but rather artistic realism, probably best depicted by s_v1.1 too - “Vivid Realism” before the camera was invented is more the gist.
Sorry for the delay on this - Shnell should be back on the menu in the coming week or so!
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.













