⚠️New updates to Flux Reality will be posted to Patreon ⚠️
V11 - What's new:
This is a softcore only version with no hardcore concepts trained into it.
Boobs got a big upgrade, lots of new shapes and sizes added to the dataset, this improved the nipples compared to the previous softcore version. I will expand it even further.
Pretty major stability improvements with this version
Some of the example images were made using my amateur photography LoRA trained on v11: https://www.patreon.com/posts/amateur-style-133959021
Concepts:
Big ass - Tag: big ass
G-strings - Tag: g-string
POV Cowgirl - Tags: vaginal sex, cowgirl, breasts, man, pov, spread legs, male pubic haie, penis, girl on top, straddling, from below, solo focus
*Example prompt: A 21 year old asian woman with long blonde pigtails and medium breasts looking at viewer with an open mouth nude having vaginal sex with a man in the cowgirl position with spread legs squatting from POV from below with male pubic hair penis, pussy. girl on top, she is adorned with a black choker and she is straddling outdoors. female pubic hair, realistic, cowgirl position, solo focus, upper teeth only, she is riding on top of him, translucent white cum dripping out of her pussy and down his penis, crotch focus, full body shot
POV Reverse Cowgirl - Tags: vaginal sex, reverse cowgirl, ass, pussy, anus, male pov, man, penis, male pubic hair, from behind
*Optional tag: squatting
*Example prompt: Woman, 21, with long hair, large breasts, and black hair, adorned with jewelry and nail polish on her fingernails, looks back at the viewer with an open mouth while engaging in vaginal sex from behind indoors on a bed. Her ass is prominently displayed, as are her pussy and anus, while the mans penis male pubic hair and testicles are evident. Elements like a bracelet, lipstick, and a mole accent her look, with the scene set by a backlit window with blinds, rendered in a realistic reverse cowgirl position from a male pov.
POV Boobjob - Tags: male pov, paizuri, breasts squeezed together, nipples, penis, solo focus
*Example prompt: Caucasian woman, age 24, medium breasts, looking at viewer, long wavy platinum blonde hair, large breasts, male POV, brown eyes, nipples, penis, solo focus, nail polish with matte sapphire finish, long blue nails, eyelashes, paizuri, freckles, breasts squeezed together, realistic, majestic mountain range, cum on her breasts and dripping down his penis
Anime Style - Tags: Anime woman, anime style
*Not always the cleanest output but it can be fun to play around with. Sometimes it makes anime/realism hybrids, other times the anime style is more dominant. I didn't train this into the model this was already trained into Flux Dev but it wasn't very usable until this version.
Blowjob / Deepthroat from side - Tags: nude woman performing fellatio to a man view from side, male pubic hair, testicles, solo focus, deepthroat, penis, tongue out, fellatio, close-up
* Optional tag: looking up
* Example prompt: A nude woman with eyes closed while performing fellatio to a man view from side, she has long, lustrous copper hair flowing in an elaborate style with twisted, ornate braids, wearing jewelry with earrings and a choker in a realistic outdoors raw photo 8k, male pubic hair, testicles, solo focus, deepthroat, penis, tongue out, fellatio, close-up
Facial - Tags: open mouth, tongue, facial, penis, cum, solo focus, cum on tongue,
* Example prompt: A woman with long hair, looking directly at the viewer with brown eyes and adorned with jewelry including a necklace and earrings, shows an open mouth and tongue as she is engaged in a realistic facial moment; her brown hair frames her face while a nude man's penis provides cum, cum in mouth, and emphasizes his solo focus with a cum on tongue detail, the mans penis is ejaculating cum on her face
Ahegao face - You can start the prompt with "a woman making ahegao face with her tongue sticking out and her eyes crossed" or "A woman making an exaggerated ahegao expression—with her tongue playfully sticking out and her eyes crossed." Or use some variation of this in your prompt.
Innie pussy - No tags needed, just use "nude" or "pussy".
Missionary position - Tags: missionary, vaginal sex, male POV, spread legs, lying, pussy, penis, hetero, solo focus
*Example prompt: Woman lying on her back on a bed in a missionary pose having vaginal sex, bathed in light shining through a window, captured as a raw photo 8k. From a male POV, she looks directly at the viewer with an open mouth and eye liner. Her hair is auburn and she has large breasts with visible nipples. Completely nude except for emerald striped thighhigh socks, she presents a sensual scene with spread legs that reveal explicit details including her pussy. The realistic composition also features a penis, emphasizing the hetero encounter with a solo focus on every intimate detail
*You can make the penis bigger by prompting "big veiny penis", you can also add further description to the man such as "male pubic hair", etc.Doggystyle position - Tags: doggystyle, vaginal sex, male pov, pussy, anus, penis, male pubic hair, solo focus, hetero
*Example prompt: Woman on all fours on a bed doggystyle position bent over top down bottom up having vaginal sex with a man, bathed in light shining through a window, captured as a raw photo 8k. From a male POV, she is facing away with head turned to to the side with an open mouth and eyeliner. Her hair is sandy blonde, long and wavy, and she is completely nude except for rainbow striped thighhigh socks. She is skinny with a big ass. Pussy, anus. The realistic composition also features a big veiny penis and male pubic hair, emphasizing the hetero encounter with a solo focus on every intimate detail, view from behind and above.
Cum - Tag: "cum". You can specify where and how much of it you want. For example: "excessive amount of cum dripping off down penis, cum on her face and dripping off her tongue" etc.
POV Blowjobs - Tags: "woman, looking at viewer, man, hetero, nude, penis, solo focus, lips, pov, erection, oral, fellatio, male pubic hair, realistic, pov crotch, male pov" It can help to also include how you want your model posed like kneeling, lying on her stomach, on all fours etc. You can also include the view angle like 'from above' and 'close up' if you want.
Pubic hair - Tags: "pubic hair," "female pubic hair," "landing strip pubic hair," "male pubic hair," or "excessive pubic hair" for even more.
📌 Check the pinned image posts for prompts to get started. Another useful technique is using a LLM to incorporate danbooru tags into natural language prompts. Many of my example images were generated that way.
LoRA's:
LoRA's are compatible with this checkpoint if they are trained using this model (Flux Reality) as the base model.
LoRA's trained on this model work with other Flux Dreambooth fine tune checkpoints.
LoRA's trained on Flux Dev are not compatible they will burn the image!!!
LoRA's trained on previous versions of Flux Reality work ok with this version but retraining is recommended to take full advantage of the new features.
LoRA's trained on softcore versions of the base model tend to work better with other softcore versions, and LoRA's trained on hardcore versions of the base model tend to work better with other hardcore versions.
Sampler:
Preferred sampler is dpmpp_2m with sgm_uniform scheduler with 35 steps.
Prompting:
The model was trained using danbooru tags (except '1girl' was replaced with 'woman') but that doesn't mean you have to solely use them. They can be incorporated into natural language prompts which works best. You don't even have to use them if you don't want to. You can prompt in an entirely different language if you want. I tested French and Italian.
Use "realistic" in the prompt for a DSLR photograph effect. Use "amateur photo" to get a more amateur quality. You can also use my Amateur Photography LoRA for Flux Reality which is hosted on Patreon.
Poses:
To get more complex poses that weren't trained into the model use Canny ControlNet!
Workflow:
If the workflow isn't attached to the images you can get it from the #faq channel in my discord server https://discord.gg/UFkPTfTDWs
Description
Significant stability improvements and reduction in artifacts with this version!
Like V2, this version improves nipples, pussy, ass, faces, and gives outputs a highly realistic look,
FAQ
Comments (62)
Doest it work with forge ?
Yes it does, i use forge and it works for me, but you will need other files to run it, AE, Clip_L,t5xxL_fp16 or fp8, depending on your Gpu Ram,
@DxDRG 24GB 3090
I'm pretty new and using flux1-dev-bnb-nf4-v2.safetensors in the basic pinokio Forge install, that works basically alone without clip etc. I had troubles using AE, clip with other checkpoints, maybe wrong versions of everything. i must give another try as it look more promising and powerful... Thanks for the anwer !
same GPU as myself, i got my ae, clip_l and tp16 from hugginface, you'll figure it out in no time, atb
i'd recommend you also try this on other styles, it works really well,
What are some of the keywords you are using to prompt these styles? I will try them out
This is genius, I generated a NSFW picture of a woman in the woods and it's mostly dark.. I guess it's because she's in the woods :)))
Hello.
I´m trying to create a Lora for this Checkpoint with Fluxgym and I need to figure out how I can choose this Checkpoint as my model. To choose it I have to edit the yaml file but I don´t know what I have to put in "repo" and "base"
custommodel:
repo: ?
base: ?
license: other
license_name: flux-1-dev-non-commercial-license
license_link: https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md
file: getphatFLUXReality_v30.safetensors
I hope someone can help, thank you
no idea sorry kohya ss is all I use
@getphat Thanks I´ve tried it with kohya and it worked perfectly
I put another model then I modified the batch with the right model and it seems to be working, we'll see the results in a bit
@marcomazzarella90878 I had very bad results at the end with Fluxgym compared to Kohya_SS. The results are much better then I have expected, also thanks to this tutorial: https://www.youtube.com/watch?v=nySGu12Y05k&t=1s
@bennyjohn187 how many epochs? I wanted to try it too but I saw that it does training with 64 GB of RAM if you have a configuration file I'll take a look
Great! Any chance that it will work on Lora’s trained on Flux 1D in the future?
No, as a community we need to make better checkpoints and leave flux dev behind, our lora's will be cross-compatible then.
@getphat Hopefully :) thanks.
@baablebrox I've tested this already, I've trained LoRA's on Flux Reality and they work with other peoples' flux dreambooth checkpoints, the incompatibility is between dreambooth checkpoints and flux dev
i want to train a personal character lora model with this checkpoint but Civitai's on-site trainer does not show it in the list ?? In fact, only the basic Flux-Dev seems to be available for training yet its unable to draw a damn nipple......
Simply amazing, I'm having too much fun! Understands natural language so I can feed my ideas to chat gpt and fine tune the results. Love it! Can't wait for v3 to be available.
.
Grab the workflow from this pic and drop it into comfyui to load the nodes and settings
.
Only the chest looks normal. Everything below is, to put it mildly, crap. Sorry....
You're probably going to have to use a vagina lora, even still to this day I need to use vagina loras in SDXL or they turn out bad. And SDXL is at a super refined state now.
first v3 lora trained and it's looking great. any chance of you releasing a compact version? they run much faster on my setup
Cool, these models take so long to train and test, v4 will likely take 2 weeks in all, I dont really have time to do more than fp16 and fp8
I must say right away that I do all the generation in FORGE and on RTX 3060 (12 GB). And to be honest, I have a double impression about version 3.0. everything seems to be almost fine with NSFW, but there are much more artifacts with limbs, as a percentage of the DEV version. There are also questions about the picture itself ... it's kind of loose and seems to be all blurred, and probably because of this, I didn't notice much difference between the 22 GB version and the 11 GB version, in the pictures that result.
and one more thing, I tried my LoRA models made on the DEV version - they work, but the similarity with the character gets a little worse.
and one more thing, the picture itself is much better on the DEV version, if you use the LoRA model with NSFW, it is on the resource, but I will not advertise.
Im well aware that there are artifacts with some prompts and seeds, it says so in the description of the model
The sharpness and coherency can vary a lot using different schedulers, was so with SD and still holds true with Flux and is not the same for og dev vs derivates. Would have been interesting to know what settings you have been trying.
i neeed some help to run this checkpoint with comfyui, i use the fp8 version but the image gets burned or when i use a cumzilla lora not work with fluxgym lora. can you help me?
Reduce the weight of the Loras and train the Loras off the checkpoint.
@Kierkegaard420 i will try. thanks..
lots of anatomical anomalies, but when it goes well the quality is good, still far away from the best SDXL ones still.
What model can u reccomend me in SDXL?
yup thats why I just spent 9 days training v3.1 using regularization images which fixed the stability
@Momongas All rounder model, too many to list, but NSFW specific my current favorite is Lustify
Hello, How to use this checkpoint correctly without burn image problem with ComfyUI? Any workflow to work with this checkpoint? Thanks in advance.
Just grab the workflow from one of my pics that has it, save it and drop it in the comfy window
Has anyone trained a lora using this checkpoint? Does anyone know how to point ai-toolkit to this checkpoint for lora training???
Default line item is:
name_or_path: "black-forest-labs/FLUX.1-dev"
Thanks
ai toolkit work only on flux dev
@volviqe438 That was my thinking, thank you for confirming!
How to install it in Drawthings app on M4Pro ?
I was able to import the 3.0 FP8 edition just fine. I had already installed the Flux [dev] 8-bit from the models menu, so that probably helps.
Can this model also be run in Forge WebUI? Because I get an error like this:
AssertionError: You do not have CLIP state dict!
This model is just the unet, you'll also need the clip, text encoder, and the vae.
Lots of Buzz wasted.
@Subseven8888 Not really....just go download them....
https://github.com/lllyasviel/stable-diffusion-webui-forge/discussions/1050
@Nexvus THX! That's all I need it. Seams to work well in Forge.
it works fine!
Seems like the data set contains a significant number of images with hands hidden, which is visible with the bias towards such images, which has the effect that the model has less opportunity to learn hands which then leads to a reduced quality. Kind of an old mistake i suppose that was introduced because of troubles with poorly trained SD models. Kinda like what you can't see can't be bad... :D
Also the usage of WD14 danbooru on a Flux derivate is at the least unfortunate.
And this is disappointing in combination as it seems the images would have much more potential with proper tagging and a pinch natural language.
I know that it is real work as I am working on a dataset myself.
The reason for my critique is this, tags only works only with the loss of contextual control. Why using Flux in the first place if the ability of its text encoder gets so little to work with? It is a waste of resources to be frank. Especially because the computation on this model is so freaking expensive, it is damn worth it to invest much better in the data that goes in.
That is a damn shame about using the fucking danbooru shit. The great strength of Flux is being able to prompt in natural language. I probably wont bother trying this model.
can anybody tell me how can I do a reclining/arched back pose while lying on back? It's impossible to do with flux.
canny controlnet
what's new in the version updated 24 days ago?
its the same version, but I got a new one coming soon
Can you link your workflow?
Gladly minimal solution for beginners
Hi can I ask a workflow to make a lot of faces of the same model? Im lost to how to make that happen :(, I wanted to build same character but idk if its better to "faceswap" or just build a lora*
train a lora using this model as the base model instead of flux dev
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.



















