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
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
FAQ
Comments (57)
oh no can't get early access all my buzz is gone because of Flux
time to wait I guess xD
You have enhanced this model and so many of my other models so much with your great work. I would love to invite you ♥
@AIDigitalMediaAgency thank you so much😍😍😍
I received a lot of buzz lately, I gave you some
You're telling me this model can somehow get better? Haha congrats on 1.0!
haha thank you, tried my best. Hope you like it ♥
Many thanks to everyone who uses ForReal and walks this journey with me. ♥ ♥ ♥
This is my biggest and favorite project. Without ForReal I would not have been able to upload 80%+ of my pictures.
I'm looking forward to V2.0 but please don't ask for a date 🤣
My hidden Message for you! ♥
I know exactly how you feel, huge thank you for what you're doing to the community!
@HamedEmine thank you so much ♥
I would like to see a bf16 or fp16. :-)
will release it in the next 24 hours
@AIDigitalMediaAgency cool. thx very much.
I still do net see it listed :( Can you recheck your upload please ? Thank you for the great work!
has simply filled the early access target on its own. Please thank him for this abnormal support.
Thank you from the heart ♥
you are crazy
Awesome,finally The Big one, Will You uptade on tensor?
i will.
@AIDigitalMediaAgency The goat
Where I can get DDPM sampler?
In ComfyUI or SD forge (better A1111)
When people say their new model is better than their last, I always take it with a pinch of salt. Normally, it’s over-trained and not as good, but this is a great update. Great work you know what you are doing!
Thank you so much ♥ Glad you like it!
v1.0 is great.
Time to bring your talents to Flux!
when do i use this over flux and whats a comfy workflow i can use for this
You use this model over flux when you want to use this model. Any comfy sdxl workflow can be used, the recommendations are listed. User name checks out.
@threeRthreeDe ??? no need to be so rude
Hey!
This is a SDXL Model, this is not a FLUX Model. If you want to use Flux and Flux Loras, you can use my simple Workflow for FLUX: https://civitai.com/models/618766/flux-comfyui-with-lora-support-txt-2-img-prompt-variation-without-upscale-for-small-gpu-and-ram-fp8-fp16-beginner
If you want my ForRealXL Model, you can search in the Model Tab and filter for Workflows and SDXL. There you can find a lot of workflows to use this ForRealXL.
You can also use A1111 or SD Forge with ForRealXL.
@AIDigitalMediaAgency i was asking when to use it OVER you might have misinterpreted my sentence /nm
@dontknow_ Okay, I don't understand that. What do you mean?
@AIDigitalMediaAgency like, with what tasks should i use thi with rather than flux?
@dontknow_ Both SDXL and Flux are great at a lot of the same things, so it is a matter of preference. SDXL and Flux are nowhere near as different from each other as they are from SD1.5 and Pony.
The biggest differences:
-SDXL has some built-in knowledge of artists
-Flux has better prompt adherence
Both are great.
Use this with SUPIR upscale over flux. It will do it faster and be excellent quality. Check my profile for a comfyui workflow.
How do you get this model to draw full body shots, from head to feet? Both .5 and 1.0 I can't get it to draw knees or lower.
scroll through the galery an check some of my settings. use some words like "full body, full body shot, full body photo, ..."
@AIDigitalMediaAgency Yeah, that's what I am doing as well. (((full body shot))), zoomed out view, head to feet, and so on.
Is there perhaps a specific W/H you trained on, and use before upscaling?
@civtai25 no you should be able to use any ratio. i think it's more of a typical sdxl problem that it's harder to get the whole body. but it is possible. I don't have any examples, I would have to browse through my pictures myself, the last time I used sdxl was before flux release.
is there any plan on making inpainting model?
Unfortunately, this is not planned
inpainting models isnt needed anymore with SDXL. you just need a specific comfy node.
@AIDigitalMediaAgency okay thank you!
@zGenMedia i've tried a lot but all the models didn't work well.
when it comes to generating background(not small area), all the regular model sucks. Any recommendation for specific node?
Isn't there a new version of SDXL planned? Maybe ForRealXL 1.5-2.0.
I am sure that one day a new version of ForReal will be released. Please do not ask for a date 🫠
@AIDigitalMediaAgency what is the capability of learn the sdxl model 'that 6.5gb' it's unlimited ? maximum ? it's limited to what ?
@amazingbeauty I wonder the same thing.
This is one of the best SDXL Turbo/Hyper/Lightning finetunes I've tried! I'm surprised that it isn't more popular.
From comparisons I've seen and my own tests, both DMD2 and Hyper outperform Turbo and Lightning by a wide margin. Changing the base model could make this model even more capable. Do you have any plans for this?
Again, thanks for creating this great finetune! Look forward to seeing more of your projects in the future.
Thank you so much for your experience. I'm honestly not sure where the journey with this model will take us.
So glad you like it ♥
I'm having trouble with Jugg, so i tried this. And now I'm very happy i did so. Thanks for sharing!
thank you so much, happy you like it ♥
Is it possible to make an inpainting version for this model? Does an SDXL Lightning inpaint version even exist?
Unfortunately, this is not planned
I don't quite understand why this checkpoint isn't as out there and as mainstream like some of the others. This is insanely good, If anything this checkpoint works better for the loras I train than using the Original checkpoints I trained my loras on. Even when I use controlnet it doesn't lose its sharpness and quality like other models. Very nice.
Hi, can you provide a GGUF version of your model? thx a lot for thise great model
ComfyUI unless instructed otherwise will probably load it in fp16, so same size as any other model.
GGUF can be made.
bro please we need new model
It's 2026 and this model is stil awesome!