Hello ♥
for whatever reason you want to show me appreciation, you can: ❤️Ko-Fi❤️
Special Thanks to
badassdragoon2 georgebanjog Mr_fries1111 barytdruck124
Please take a look at their amazing work!
Update v1.0:
if you see when V0.5 was released, you can imagine how hard it was for me to create a better version that is really better in all areas.
This is the result of probably 100 different versions.
more Details
better colors at low cfg
better faces (and more variance) (not perfect)
For people and faces, you should not use DDPM. Use DPM++ 3M SDE or Euler
Update v0.5:
Small improvements
- better skin, teeth and eyes in humans
- more details
I am still working on the following improvements
- deformed faces / displaced eyes
- faces from a distance
- xxx
Update v0.3:
It's just a small update to fix some issues from v0.1.
It doesn't change every subject, but some improve.
I haven't had a picture in about 1000 pictures where v0.3 wasn't better, so the upgrade is definitely worth it.
I can notice the following improvements:
a bit sharper
more realistic
a bit more creative
a bit more possibilities to get nicer hands/faces on full body photos
________________________________________________________________________________
This is a merge of the following Models:
Boltning
JuggernautXL
WILDCARD Lightning
Recommended Settings:
Sampling Method: DDPM (my absolute favorite), Euler, DPM ++ 3M SDE
Sampling Steps: 10 (more very rarely make a difference.)
CFG Scale: 1-4 (lower, more realistic, i prefer 1-1.8)
Upscaler: SwinIR _4x (this is my all time favorite)
(Hires steps: 15; Denoise: 0.28, upscale by 1.5 - 4 (depends on your GPU))
📌 USE sdxl_vae
All Images generated in SD Forge with Nvidia RTX 4090
Please test other Settings and share with the communty, so i will learn about the Settings and can optimize.
Recommended Loras:
Please experiment with the weights. Loras are really gamechanging in this Model!
TESTS
Here you can find and download my tests:
Thank you so much for using this Model. Hope you are get insane results and create some BANGERZ ♥
Description
v0.5
Small improvements
- better skin, teeth and eyes in humans
- more details
I am still working on the following improvements
- deformed faces / displaced eyes
- faces from a distance
- xxx
FAQ
Comments (70)
Wow best Checkpoint Merge i've ever seen, can wait for the NSFW to be added.
Thank you so much ♥
An NSFW Model? Do you get bad results with NSFW?
@AIDigitalMediaAgency yes I had some problems with penetration and stuff like that, i always try sci-fi scenes with bioluminescent creatures ect but i can't get what i want haha. it's like the checkpoint don't want it. But for artistic and beautifull scènes it's perfect.
@Serarts yeah okay,this is more easier to get with an pony model. Maybe you can check my Pony ForReal Checkpoint, but of course not comparable.
@AIDigitalMediaAgency Yeah i will try it. Once again thank you
please upload to tensor art
I didn't know the site at all.
i will.
Thanks for the info!
@AIDigitalMediaAgency and sea art as well pleaase
TensorArt is the worst.. probably end up over there anyways when they steal it
@TheP3NGU1N lol i know,that is why i suggested sea art
@kd231002573 which is not much better either in that regard..
@TheP3NGU1N better than nothing lol
There must now be five Checkpoints on Civitai called Pony For Real, they're even showing up in the recommendations. Rebrand please! How about PONER is like Pony but with a...
yes, i've already thought about it. i'll give it some thought, thanks for your feedback
almost I try Euler-a can not work with InstantID ? T.T
you have to try Euler without A. But i dont know what is InstantID.
Same for me, only Euler A works. Results are too grainy no matter what.
Your previews are over-saturated of LORAS. 🤔 Not a good way to showcast the real potential of the model.
There are very few loras in the images of the model preview. This checkpoint works great with a lot of loras, especially in combination. I'm experimenting a lot here, so I don't get to take pictures without loras. If you agree, I'll pin your post at the top :)
@AIDigitalMediaAgency Such case, what if you add also some previews with no loras and pure prompts, so the checkpoint can be properly showcased? 💡
@cybermiaw there are quite a few distributed here, for the Prewiew of the model I of course take the best pictures.
ENDED
🚨 Featrure Image BOUNTY 🥇
if your picture made with ForRealXL (any Version) appears in the featured pictures, I will donate ⚡️ 1000 Buzz ⚡️ to the picture. Good Luck
Please send me a message or comment here if you have won.
Thank you for the wonderful model!
The following image has been featured in "Featured Images".
@aihonobono2023 congratz! ♥
@AIDigitalMediaAgency Thank you very much!
Guys, this Event is over. Congratz to all! ♥
Hello Friends,
Before I download this model, I'd like to ask everyone how this model generates vehicles. Is anyone satisfied with it? You know, it's almost 13GB, so I want to check first. I don't want to give my PC a hard time :)
I've already taken some cool vehicle pictures, you should be able to find a few here, maybe in the older versions too.
https://civitai.com/images/13359849
https://civitai.com/images/12066066
https://civitai.com/images/13514029
https://civitai.com/images/13462229
download the fp16 version which is half the size
@hdmpkt I love it !
@Tumi007 Its good idea. Thanks !
Glad you like it! Nice Results ♥
Guys all your images are freaking AMAZING!!!
Wow, I can hardly keep up with looking at them all at the moment. Thank you so much for enriching this model so much! ♥
i wish i could globaly tip you all together, i think i'll suggest that in the next stream.
After a lot of test ,This model is my best model for me right now. It cover all of prompt i need and give a very clean beautiful image. Good job keep it up brother.
thank you so much ♥
Your Images are insane!
Hope we will se a lot more of your nice Masterpieces=)
Hi! How can I get the DDPM Sampler??
I'm using Automatic1111
Upgrade to SD Folge, its an upgraded A1111
@AIDigitalMediaAgency Thanks! Where can I find SD Folge??? There is no info on the web
@santiagopgm8 Sorry, SD FORGE
Can you allow civitai generation?
i can't decide that, civit decides which models are released.
I have allowed it, as soon as Civit allows it, you can use it.
is there anyway to ask about it?
@daldassak7681 i don´t know the requirements. I don't know who decides that either
Hello. How can I use this model on the Fooocus platform. I have tried all methods. I can't get a healthy image output.
nice model my friend
Thank you so much ♥
Hi, I'm generating via ComfyUI and have encountered a rejection of the specified most interesting LoRa, namely the error "ERROR lora diffusion_model.output_blocks.3.1.proj_in.weight shape '[640, 640]' is invalid for input of size 1638400
ERROR lora diffusion_model.output_blocks.3.1.transformer_blocks.0.attn1.to_q.weight shape '[640, 640]' is invalid for input of size 1638400
ERROR lora diffusion_model.output_blocks.3.1.transformer_blocks.0.attn1.to_k.weight shape '[640, 640]' is invalid for input of size 1638400" and so on for each LoRa. Perhaps this is due to the 32 bit model and I should download a 16 bit one?
Thankies a lot again for your help 🩷 Your checkpoint definitely works wonders ✨✨✨
Your Welcome. Thank you so much ♥ Hope we will see a lot more of your amazing work =)
what specific wonder it can make or would make ?
I am so incredibly grateful for everyone who uses this model. Your pictures are so incredibly good. We have managed to get more than 30 images featured with ForRealXL. You are all amazing!
Thank you so much 🫶
one of the greatest lightning models, Congrats on your achievement my friend! ♥
@DreamCk thank you for all your amazing images ♥
Thank you for this cool model ! And congrats for all the features !
Thx you for this amazing model ❤️ I hope that many more featured images will follow 😉
Amazing model !
Thank you !
I'm so happy to say that the issue LoRas with keys of weight was solved by switching to SD Forge! Everything is great, but if anyone needs an emphasis on NFSW content, I highly recommend the model PonyForReal from this author. The author is very hot and has great models. I thought I'd be unimpressed by now, but the subtle integration of LoRas that other runtime environments can't count without errors works wonders! I thought I would never move out from RealVisXL, but it happened! I am pleasantly shocked 🖤
Thank you so much for your kind words. glad you like it. Yea, SD Forge is way bettere than A1111.
Hope you get a lot of Bangerz ♥
nice
I was very excited to use your checkpoint and attempted to implement it into the Foocus UI, using a Reddit link in one of your previous comments: "https://www.reddit.com/r/fooocus/comments/1b25rd0/how_to_use_lightning_models_in_fooocus/".
However, the images were coming out very grainy attempting to recreate one of your other masterpieces of a haunted house. I thought maybe it was the LORA's, but removing them didn't help.
I have some experience using A1111 and was able to get the model to work, but my results were more cartoonish than I wanted. Might need more practice.
I switched back to Foocus to see if I could get other lightning model (juggernaut V9) to work using the same settings. The images were clear. Not sure why I cant get your model to work. I see you're using forge from other comments, but I don't belief using different UI shouldn't alter results seeing they're playing with the same settings..
You're work is very impressive!! looking forward to seeing more of it now that I'm following you.
Edit: ForRealXL version 5 works on Foocus!!
Thank you very much for the experience you have shared ♥
I am of course only new to Stable Diffusion and don't understand a lot of things, but according to my promt this model works better on "DPM++ SDE Karras" Steps - 6-8, CFG Scale - 2.
For my fantasy world requests these settings are ideal, and this model surprisingly fits best of all. Thank you very much to the author! You are great.
Yeah absolutely, the settings and desired results are different for everyone. I'm glad you've found the perfect settings for you ♥
I also got a better result with these settings
harika bir model