Join me on Patreon for exclusive perks and early access to unique resources.
To discuss custom LoRa's or models, feel free to connect on Discord.
Meet ULTRA_REALISTIC_BY_STABLE_YOGI
Why a new model?
I fused the strengths of my two flagships—Realism_By_Stable_Yogi (hyper-detailed textures) and Babes By Stable Yogi (stylistic “babes” aesthetic)—to remove the hassle of juggling checkpoints.What’s in it for you?
Single download, double power: One drop of "Ultra Realistic by stable yogi gives you the photoreal skin, fabric and lighting of Realism plus the flattering poses, proportions and vibe of Babes.
Cleaner prompts: No more switching models or stacking LoRAs; just load, prompt, and render.
Consistent outputs: Both styles are blended and re-balanced so results stay sharp and coherent across various seeds and resolutions.
Bottom line
ULTRA_REALISTIC_BY_STABLE_YOGI lets you tap the best of both worlds—studio-grade realism and the polished “babes” look—without extra setup. Grab one checkpoint, create more.
Description
Pro Version of most of my models are available on My Patreon
Sampler Euler a, DPM2 a, DPM++ SDE, DPM++ SDE Karras, DPM++ 2M SDE-- SGM Uniform,
CFG Minimum 4+ maximum upto 7
Steps Minimum 18 , Recommended Steps 27
Resolutions: All SDXL Resolutions
Adetailer (Required)
High res Fix (Optional)
Denoise - 0.30
Minimum hires steps 5+
Minimum Upscale 1.5
Upscaler 4x-UltraSharp
Keep the VAE at Auto ( VAE Included )
Negative Prompts
(Download this negative embedding for best results) Stable_Yogis_PDXL_Negatives
Positive Prompts
(Download this positive embedding for best results) Stable_Yogis_PDXL_Positives
FAQ
Comments (34)
Apologies if you're already planning on this, but could you upload a non-accelerated version of this model? <3
Yes uploading
Can you please upload SDXL version?
Sure, will work on that. Illustrious and SDXL will be coming soon. Thanks
Thanks again for the amazing work mate! Do you have any suggestions on the settings to use to train a lora on it? My tests with koyha are not working for now
Thanks for the love and feedback! 🙏
Use fp16 for training, choose the standard presets in Kohya
This (ultrarealisticv10 fp16) works with ipadapter, if anyone having problem with ipadapter and this model, try remove clip_vision and insight_face (they are both optional, so its weird that omitting them make the node work better, duh). (i've already play with faceid weight until it genereated distorted face). But damn, this one is definitely a keeper. Superb details, great texture, and so on. Most notable detail that i really like is the eyes and skin. texture. Thank you for this great piece of art. i have yet try the other models, but i think there should be no more problem regarding ipadapter.
Thank you for testing and feedback, means a lot 🙏
I thought it was just me. Can confirm the same. Reproduces some of the elements - renders them incoherently.
@waaz1d383 You are talking about the DMD2 or Fp16 from version 1.0 ?
@Stable_Yogi actually i'm a fan of your work, so i can say that all of your latest release (babes, realism, asian realism, pony, ilustrous, \\with exception of SD1.5 models, iirc they work well with ipadapter,\\) from mid 2024. Most of them seem to have "strong" face that render ipadapter unable to perform well. I just don't have enough space to experiment further, but will do so in near future.
Edit: it maybe from comfyui though, tried forge and fooocus, but can't make heads or tails with those. Actually i don't even know how to work with A1111 too, maybe i like to complicate things... lol
@menyan1381 Thanks for the love and support. I do most of my images in Forge ui and haven't faced such issues yet. But i will look into this for sure.
@waaz1d383 try removing insight_face and clip_vision, those are the culprit. i'll soon post my workflow
@Stable_Yogi Would you say that Forge is better than Foocus to get the best results with your checkpoints? Or it doesn't matter which one? I'm learning and still trying to understand the best workflows.
@anthonirox I have seen Forge doing better. Focus is limited.
@Stable_Yogi sorry for late reply, It was the Fp16..
@Stable_Yogi @menyan1381 thanks, you're correct. I use an IPadapter FaceID node and a ClipVision enhancer.
@menyan1381 this is what it means to be a great community contributor. Stable_Yogi's work has been the metric by which SDDXL/Pony models have been measured. Its good to see such positivity and communication flourish on the platform - especially supporting the people doing the heavy lifting. Kudos to you.
Seems like this can't inpaint like Realism can?
Yes, its less effective in in-painting., Realism does it nicely. Will add in next update.
Dear Mr. Klingum_AI I kindly ask you to group your works at least 10-20 pieces. There is no way to view the works of other authors. This is some kind of advertisement for a store, with the same model. A completely banal plot, and the same type of images(( Thank you for understanding.
lol, i agree
This is one of the best models around, kudos.
the V1 DMD2 incredibly realistic
I must say with a few LORAs you can achieve great realistic images with this checkpoint. Keep up the great work Yogi.
Thanks for the motivation. 🙏
@Stable_Yogi posted a few images showcasing this
Hey thanks for your amazing checkpoints, your work is astonishing. I was just wondering why this wonderful model is no longer available for onsite generation. Keep the good work 👏
Thank you !
We have to bid for the checkpoint to stay available for image generation. this has to be done every week. https://civitai.com/auctions
I try my best but don't have the buzz to fund all of the bidding by myself. Kindly bid and keep it above in the Top 150.
Thanks
Does anyone know how to make the same effect of a clear background without blurring as in the examples?
"Focused background" or "background in focus" within your prompt. Alternatively, "blurry background" in your negatives.
@Altra Thanks! In most cases it doesn't work, but with some checkpoints it works well
shallow depth of field or DOF in negatives. possibly try "split diopter" in positives but I haven't tested that one yet
Tryin 2 use this checkpoint which is a fabulous 1, really good realism, but Civi's generator sayz "Some of your resources are not available for generation: v1.0_FP16" and it will not allow me 2 use it, even tho it is listed in available chkpoints. ???