Please consider buying my two albums Sisters of the Wicked and Danger Ahead, my book Dracula and also please check out my YouTube channel. Civitai won't support creators, so any purchase is a real help. Thanks.
DRACULA https://www.amazon.com/dp/B0FM4DXYPL
Sisters of the Wicked https://amazon.com/music/player/albums/B0DSJY3ZHY?marketplaceId=ATVPDKIKX0DER&musicTerritory=US&ref=dm_sh_CdyeuxIBNBDOjZ1hpB1PPJxT0
Danger Ahead https://www.amazon.com/music/player/albums/B0DB2DV354
Music In The Machine https://www.youtube.com/@musicinthemachine/videos
Lora Creation: https://www.upwork.com/services/product/design-flux-lora-for-creation-of-ai-art-1909718528813429477?ref=project_share
For the De-Distilled Model, Distilled CFG Scale should be set to 0. Then use actual CFG. For me 3.5 to 8 seem good.
If you like the models, please follow. The page is getting crowded as I keep releasing my model in as many formats and options as I can.
V1.1 Hyper 8-Step is now available and works in ForgeUI as well as Comfy. Schnell V1 Much faster, getting good results with 1-4 steps. Quality is not quite on the level of the Dev model, but much faster. So far it seems text is more consistently good though. New GGUF version of the Schnell model up now. Thanks to pkmngotrnr for converting them for me. Much appreciated. So far works in Comfy only at least for me. DEVfp8V1.1GGUF is now up as well. V1 version has European Flux mixed in at what seemed like a pretty low percentage, but it seems like it is a bit too strong. If you like this look, feel free to use this one.
This is my initial FLUX model. It is base on the flux1-dev-fp8 model. This is a merge only in the sense that I merged in some Loras that I like to make it more realistic. As with all of my models, the goal is the best photo realism that I can achieve. If it will do other things well, that is a bonus. Until I can learn a lot more about FLUX, and see if at some point true merges are possible, this is a pretty good start. I don't know about compatibility with all software. I use ForgeUI and I am able to run it fine. Most gens take about a minute and a half for me at 1024 X 1328. I use Euler with the Beta scheduler or the Simple. I haven't experimented much yet with other samplers. I typically use 24 to 28 steps.
Description
GGUF 4_1 This version incorporates my Acorn Improved Nipples lora into the model. I can't promise it will entirely eliminate the dreaded FLUX nipples, but it does most times. It is also less prone to give you nipples on the outside of clothing as some of the nipple fix and breast loras tend to. I have found that on occasions that it does, you can apply the Acorn Improved Nipples lora with negative weight in the prompt and it usually stops it. Also the lora works great any other time you get nipples at unwanted times. Apply it like <lora:nipples:-0.99> in the prompt.
FAQ
Comments (7)
Sorry, flux noob here. Downloaded the most recent gguf file put it in the same folder I had flusxchnell in for forge. I tested flux last August and haven'treally used it since. I had downloaded a separate forge UI install just for flux, so this version of forge I am attempting to use is from September of last year. Just wanted to ask before I spend a bunch of time troubleshooting/updating/reinstalling, can this newest version work in forge at all? And if so will it work in a version from last September? I'm happy to do a new install of forge if needed, but am short on time most days so would like to avoid if unnecessary. If you're unsure if it will work, could you let me know what model folder the gguf file should be in? Thanks!
Updated forge and followed this guide and everything is now working flawlessly. Thanks for the model!
https://civitai.com/articles/7828/simple-setting-forge-ui-for-runing-flux-no-include-vae-and-clip
@X0l0t0l Great to hear. Yeah, Forge is all I use most of the time. All the test images on here are in Forge.
Can you create a hyper 8 v1.5 NF4 model? My 4060 GPU takes 2 and a half minutes with the latest released model v 1.5.
This model : https://civitai.com/models/638187?modelVersionId=819165 takes only 40 seconds with the same steps and the same resolution.
Also why don't you create the all-in-one model? clip - T5 - VAE I think it takes much more time to generate.
Thanks for your time.
That's curious. My 3080 Ti only takes 58 seconds to generate at 1024 X 1280. I will look into an nf4 version. I know I made an nf4 of the previous model. I will just have to remember how I did it. I don't plan any all in one models. Pretty much everyone already has the clip, text encoder and vae already and generally don't want to waste the extra space of storing the redundant info.
The NF4 Hyper 8 Step version of V1.5 is up.
@Seeker70 Thanks a lot, yes I think the 3080 TI is better than the base 4060 8GB VRAM
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.