Aesthetically beautiful futanari and traps.
TR5.8R (more poses (masturbation, sex) more retro, more realism).
TRW1.56CON Boosty (NEW VERSION - added chastity cage, condom on penis, greatly increased dataset, diversity and stability)
P.P.S. new model Anime Screencap / LORA / PONY DIFFUSION and NEW ERA 3.0 (MODEL) PONY DIFFUSION / SDXL on Boosty
For many arts, combinations of my lore were used, such as: Esthetic 90s style, Esthetic Urushihara Satoshi, Esthetic Lina Inverse (but new version from boosty)
I made a video on how to achieve the same quality or just repeat my art
Lora trained on almost 700 images with manual prompts settings. Few big cocks, mostly medium and small ones in the sample, because models like to make house-sized cocks, I don't like that.
changes in version TRW1.4CC:
added chastity cage
The dataset has been more than doubled
the quantity and quality of anime art has been increased, which should have a positive effect on the “anime” style
heavily reworked prompting
Eta noise seed delta 31337
Use model VividOrangeMix1.0_NSFW
or Abyssorangemix3aom3_aom3a3.safetensors model
or aom3a1
VERSION 5.0R much better with low weights. Use a weight around 0.2-.0.7, 0.8 and up is too "stylized" and too one-sided. Below 0.4 often makes a pussy, but you can try and add some basic key prompts: penis, futanari, otoko no ko, female focus, frenulum, small penis, large penis (carefully, but work), medium penis
flaccid, phimosis - helps with size reduction and do what they say
Negative promts: artist name (necessarily, most of the art with watermarks), (worst quality, low quality, extra digits:1.4), multiple penises, multiple hands, multiple testicles,
just use hires fix as much as your video card is capable of, so as not to give an error, with 8gb of video memory it is 512x768 with hires fix 2, upscaler Latent (nearest-exact) or just Latent (if there are problems with the image), the picture will be much more detailed.
CFG Scale - 5
I used arts mainly: exlic, tofuubear, personalami and many many others
The training took place at a resolution of 512x768, so it makes no sense to use 512x512
Combined with my other LORA
How to use LoRA's in auto1111:
Update webui (use
git pulllike here or redownload it)Copy the file to
stable-diffusion-webui/models/loraSelect your LoRA like in this video
Make sure to change the weight (by default it's
:1which is usually too high)
*Information taken from Lykon
page on Hugging Face https://huggingface.co/OneRing/TP3.7.safetensors
Please post your work with or without comments, it will help me improve. Thanks!
If you like my work, click on the heart above, I will be pleased :3
Description
New version TR5.0 (full improved remake) on Patreon
About version 4.1 - now better in retro! This is my last attempt to teach lora into something unusual, like auto-fellatio or better rear poses, some are successful, some are not, the number of arts has been increased to almost 300, there is more variety, but the lore is already a little "burnt" , use CFG Scale 4-5, no higher. In the new version 5.0, I started everything from scratch and the results came out even better than I expected. There will be updates, so version 5.0 will be in the public domain, don't worry
FAQ
Comments (32)
Can I use this with other doggystyle Lora?
try it, the lora has back and side poses, but they are not perfect and still need to be improved
Does not produce aesthetic quality images worth producing. It produces cookie cutter spray and pray run of the mill big eye anime smut images that aren't particularly artistic or useful in context of the incredibly small pool of people who would appreciate the particular flavor. Don't cry about subjective opinions, just get better at making focused models for the audience in which they are intended.
I never thought that I would have a personal hater) That is, did I understand you correctly? You gave me 1 star because you don't like the title? But let's be honest, this is not the reason, the reason is that you do not like such content, for which just yesterday they added filters, which by the way are enabled by default and you cannot see +18 content, that is, you intentionally set the display in the filters what you don't like to put 1 star again. I ask you to put content in the filters that does not suit you, so you will protect your nerve cells from such debauchery. Good luck!)
@OneRing imagine making a whole patreon post ranting about this guy's 1-star review on your pay-walled LoRa and even a reply to it. how fragile is your ego...
@Sadolm yes, you are probably right, I need to take it easier, everyone has the right to their opinion. I deleted the post from patreon, this is really a cruel stupidity on my part.
@Sadolm I'd say I'm surprised but you know... if you know, you know. Censorship, denial of reality, etc. etc. fake reporting, canned cycles over and over if you don't lick their boot like they tell you to. What can I say, I aint skeert and I don't kneel.
@OneRing I can't make sense of this rant. If you rewrite it in a more cohesive and concise way I may be able to answer whatever it is you are asking.
So I'm using Hires. fix to upscale the images, and I've been running into some issues. My video card is 8 gb and the photo turns out fine at first, but then when it has to upscale at around 60% completion, it starts adding limbs/torsos/etc. I've used Latent (nearest-exact) and Latent upscaler as stated in the description, but both have the same issue, so I'm not sure what the problem may be. Any help would be greatly appreciated. I've also tried messing with the Denoising strength and steps.
denoising strength should be 0.4-0.5, this will reduce the changes, with a high value, the initial art changes too much. But there is a downside, if the denoising is too low, and the hires fix is too high, then the image will start to get "noise", here everything is individual for everyone, but again, the advice remains valid, reduce the denoising strength values so that changes in the art are only visual. Good luck!
@OneRing That worked, thank you so much!
@PulledPork My pleasure)
When the girl back to me I can't draw the penis.it's always just balls betwenn legs
there is such a problem and no matter how much I try, I can’t solve it, try adding promt spread legs, sometimes it can help
@OneRing ok,thank you sir
Sometimes you need be very specific in prompt. Sometimes is just luck. Try a (((visible penis between legs))), (from behind:1.4) and your stuff.
@nergaldarskiI will have a try!
Is there a way to reliably produce an ejaculating penis? There is no prompt I know of that produces semen without it flowing out of the pussy or splattered on skin/clothing. This LoRa sometimes manage to produce it. If anyone knows a way please let me know.
Since it looks like this is still being actively worked on, could you consider adding support for making testicles optional (either "testicles" in the negatives, or "sackless" in the positives or something), and add support for vaginas to have true hermaphrodites?
In fact, a lot depends on the number of arts for a particular feature. Loras, unlike models, do not really like a huge amount of art, the quality starts to drop (checked a hundred times and no settings save), I try to carefully add art, and in version 5.0, which is not yet available to the public, I completely redid everything from scratch to improve the quality, in fact, futa, which have all the genitals - are in my arts, try prompt: full-package futanari, sometimes it helps. I will try to take into account your wishes, add more full-package futanari and no testicles in future versions c:
@OneRing Thanks. Didn't realize it was such a balancing act. If it messes with the quality of the results, don't worry about it, but I hope it works out.
@neilarmstron12 I will try anyway, thanks for your request)
Sorry but I'm new, so I'm wondering since I'm not getting the same results even though I copy the generation data, we are supposed to use SD 1.5 and then activate the Lora?
use this model
https://civitai.com/models/4451/abyssorangemix2-hardcore
and the description says how to achieve the same image quality - just use hires fix as much as your video card is capable of, so as not to give an error, with 8gb of video memory it is 512x768 with hires fix 2, upscaler Latent (nearest -exact), the picture will be much more detailed. If you want a style close to the original but with more detail, don't use Denoising strength above 0.55, but to avoid artifacts, do not use too low a value, optimally 0.5
@OneRing Oh wow I can't believe I missed that, thank you very much for your quick help and the Lora :)
@Pidocles <3
https://drive.google.com/file/d/1lR7dKvFoKK9mCwKU4hVcR8jwYnYXoZgE/view?usp=share_link
@OneRing Hi, I'm facing the same problem (also new to lora) and I was trying to recreate the angel picture from the examples.
An image of my settings with my current output: https://ibb.co/k9M0j4P
Do you possibly have any idea what might be causing this issue?
I seem to be getting all art in this style (thick lines) unlike the example's thin lines and less "stylized" art.
I'm getting different results with/without lora so it seems to be detecting it at least
@uwupatty to recreate you need to know the exact size of the upscale, mostly I use 2.15, sometimes 2.0, 1.2 is too small to achieve the same quality.
@uwupatty I have accumulated quite a lot of questions under different loras, but I am now in the hospital and am not able to answer everything, I apologize
Details
Files
TP4.1.safetensors
Mirrors
TP4.1.safetensors
estheticFutanariTrap_tr41.safetensors
TP4.1.safetensors
estheticFutanariTrap_tr41.safetensors
TP4.1.safetensors
estheticFutanariTrap_tr41.safetensors
estheticFutanariTrap_tr41.safetensors
esthetic_tr41.safetensors
estheticFutanariTrap_tr41.safetensors
estheticFutanariTrap_tr41.safetensors
estheticFutanariTrap_tr41.safetensors
estheticFutanariTrap_tr41.safetensors
estheticFutanariTrap_tr41.safetensors
TP4.1.safetensors
estheticFutanariTrap_tr41.safetensors
TP4.1.safetensors
estheticFutanariTrap_tr41.safetensors









