A special blend of herbs and spices to tantalize your taste buds. These models are flux dev based models and are merged with a number of loras in an attempt to maximize nsfw photorealistic images of beautiful nude women with beautiful breasts, and with the addition of version 1.4, beautiful vaginas. While they are primarily focused on upper body portraits, they are capable of generating full body anatomically correct female nudes. Lora merge ratios were chosen to minimize bleed through when prompted with clothing. Each version is an individual mix of loras, they are not improvements of each other. Each model has its own strengths and weaknesses. A special thanks to the lora owners who allowed me to include their work in these merges. I hope you enjoy them.
NOTE: When ByteDance Hyper-FLUX Acceleration LoRA is merged into a model and the model is converted into fp8 precision, the conversion process screws-up the hyper 8step acceleration blocks from the lora. This causes the model to behave like a standard (non-hyper) fp8 model. The good news is it can still be included in the image prompt and still get the benefit of hyper 8step with fp8 model.
These are merged models of a base model flux1-dev fp16 and combinations and ratios of the following loras :
https://civarchive.com/models/667086?modelVersionId=746602
https://civarchive.com/models/681562?modelVersionId=762861
https://civarchive.com/models/652791/jibs-flux-nipple-fix
https://civarchive.com/models/660029/nakednudeforfluxsevenof9nsfw
https://civarchive.com/models/704013/photorealistic-nsfw
https://civarchive.com/models/983498/pussydiffusion-flux
To convert and merge, I used scripts from:
https://github.com/kohya-ss/sd-scripts/
To convert to gguf images, I used scripts from:
https://github.com/Zuntan03/EasyForge/
no keywords.
Description
this is just the fp16 safetensors image of v1.1 gguf requested by a user.
FAQ
Comments (24)
Need FB8 dev version, thanks.
do you mean fp8? are you not able to use the gguf version?
I have been using your masterFB8 dev version, the best fusion model. I downloaded the GGUF version and found it is not as good as the dev version, and the speed is not as fast as the dev version. I use comfui. I don't know if it is a problem with my settings. Is your experience that gguf Q8 is better than the dev FB8 version?
@xcg587568 nfsw master flux vers 1.0 and 1.1 were both fp8 precision. i have found the Q8_0 gguf to be equivalent to fp16 which should be better than fp8... if you're looking for best image quality i suggest the fp16 safetensors version i uploaded yesterday. if you can't run it or it's to slow, i can generate an fp8 version. just let me know which version you want converted to fp8?
@xcg587568 sorry if i'm confused... my NSFW MASTER FLUX fp16 (lora merged with flux1-dev fp16 saved in multiple precisions and formats) version 1.0 and 1.1 are already fp8 conversion of the original fp16 version of this model:
https://civitai.com/models/701671?modelVersionId=785079
i created it for people that could not run the fp16 version that was 23GB in size. that's why they are 12GB in size. version 1.2, 1.3 and 1.4 are fp16 precision created at the request of other users. read the 'about this version' section for each version, it describes the precision and lora ratio used to generate the models. hope this answers the question. or am i misunderstanding what your asking for?
or are you asking for an fp8 version of Flux Babes & Boobs version 1.1?
@tedbiv Yes, the fp8 version of Flux Babes & Boobs 1.1.
@xcg587568 ok, yes i can do that :), i won't be able to upload it until tomorrow.
Is the GGUF q6 also FB16 precision? It works well with my lora. The q8 makes the picture blurry. I don't know why.
@xcg587568 are you using t5xxl_fp16 text encoder?
yes, both gguf files were created from the fp16 model.
i'm testing the fp8 image now. it's about 11.6GB. i'll try uploading it tomorrow.
also i have 'diffusion in low bits' set to automatic (fp16 lora)
@tedbiv Yes.
@tedbiv My graphics card has 16G memory, so I can use gguf Q8, but the Q8 version does not work perfectly with lora.
@xcg587568 i had similar problem with lora and Q8_0 had to set 'diffusion in low bits' set to 'automatic (fp16 lora)'
ok, image is up.
id love to use this ( i use Forge UI) but no matter what , i just get this " TypeError 'NoneType' object is not iterable" , again im VERY new to AI and Forge.
have you been able to run other flux based models?
@tedbiv I run the flux dav model without any errors . Seen this and thought I'd give it ago . Then ran in to the error
@djacid my settings are vae/text encoder 'ae.safetensors, clip_l.saftensors, t5xxl_fp16.safetensors' diffusion in low bits 'automatic (fp16 lora)' sampling method 'euler' schedule type 'simple or beta'.
@tedbiv Thank you so much , I'm just about to do a test generation using one of the models showcased images, but no error as of yet. thank you for taking the time to reply and when writing this i can now see images being generated
@djacid yay! :) glad it worked.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.