Check my exclusive models on Mage: ParagonXL / NovaXL / NovaXL Lightning / NovaXL V2 / NovaXL Pony / NovaXL Pony Lightning / RealDreamXL / RealDreamXL Lightning
Recommendations for using the Hyper model:
Sampler = DPM SDE++ Karras or another / 4-6+ steps
CFG Scale = 1.5-2.0 (the lower the value, the more mutations, but the less contrast)
I also recommend using ADetailer for generation (some examples were generated with ADetailer, this will be noted in the image comments).
This model is available on Mage.Space (main sponsor).
You can also support me directly on Boosty.
Realistic Vision V6.0 (B2 - Full Re-train) Status (Updated: Apr. 4, 2024):
- Training Images: +3400 (B1: 3000)
- Training Steps: +724k (B1: 664k)
- Approximate percentage of completion: ~30%All models, including Realistic Vision (VAE / noVAE) are also on Hugging Face
ᅠ
Please read this! How to remove strong contrast.
To make the image less contrasty you can use LoRA [Detail Tweaker LoRA] in a negative value.
ᅠ
Orange Color = Optional
ᅠ
I use this template to get good generation results:
ᅠ
Prompt:
RAW photo, subject, 8k uhd, dslr, soft lighting, high quality, film grain, Fujifilm XT3
ᅠ
Negative Prompt:
(deformed iris, deformed pupils, semi-realistic, cgi, 3d, render, sketch, cartoon, drawing, anime), text, cropped, out of frame, worst quality, low quality, jpeg artifacts, ugly, duplicate, morbid, mutilated, extra fingers, mutated hands, poorly drawn hands, poorly drawn face, mutation, deformed, blurry, dehydrated, bad anatomy, bad proportions, extra limbs, cloned face, disfigured, gross proportions, malformed limbs, missing arms, missing legs, extra arms, extra legs, fused fingers, too many fingers, long neck, UnrealisticDream
(deformed iris, deformed pupils, semi-realistic, cgi, 3d, render, sketch, cartoon, drawing, anime, mutated hands and fingers:1.4), (deformed, distorted, disfigured:1.3), poorly drawn, bad anatomy, wrong anatomy, extra limb, missing limb, floating limbs, disconnected limbs, mutation, mutated, ugly, disgusting, amputation, UnrealisticDream
ᅠ
Euler A or DPM++ SDE Karras
CFG Scale 3,5 - 7
Hires. fix with 4x-UltraSharp upscaler
Denoising strength 0.25-0.45
Upscale by 1.1-2.0
Clip Skip 1-2
ENSD 31337
ᅠ
Thanks to the creators of these models for their work. Without them it would not have been possible to create this model.
HassanBlend 1.5.1.2 by sdhassan
Uber Realistic Porn Merge (URPM) by saftle
Protogen x3.4 (Photorealism) + Protogen x5.3 (Photorealism) by darkstorm2150
Art & Eros (aEros) + RealEldenApocalypse by aine_captain
Dreamlike Photoreal 2.0 by sviasem
HASDX by bestjammer
Analog Diffusion by wavymulder
Life Like Diffusion by lutherjonna409
Analog Madness by CornmeisterNL
ICBINP - "I Can't Believe It's Not Photography" by residentchiefnz
Description
The necessary VAE is already baked into the model.
This version of the model focuses on photorealism and may be worse than version 2.0 in some aspects.
FAQ
Comments (21)
Hi! Thx for v3.0, definately improved quality of photos over 2.0 but at the same time worsened similarity of LoRA for specific faces. Maybe it is problem with baked VAE? Will you also release noVAE version for 3.0 so we can test this?
Hi! I will be uploading the noVAE version of the model soon.
The noVAE version of the model has been uploaded.
@SG_161222 Thx tested it but the output is exactly the same as with VAE version, only less vibrant colors. Lora faces trained on SD 1.5 are not rendered as faithfully as with noVAE V2 that does it almost perfectly and is the only checkpoint from my testing with this ability. V3 still looks like an improvement for most use cases just not for this one. Ill stick with V2, thx for that too!
will wait for v3 no vae
Take a look here: https://huggingface.co/SG161222/Realistic_Vision_V3.0/tree/main
@diedimension1234 so which one is no vae?
For training I suppose its the way
I'm on it. I will upload a version of the model without VAE within a day.
@dogu_cat The noVAE version of the model has been uploaded.
@SG_161222 thank you <3
Thanks for including VAE!
i dont get whats the need of including vae? i already have a VAE that is automatically applyed. thanks for any explainations
@p_p People don't know what they're doing, how the mechanisms work, or what they do. So they parrot the last thing they heard and you get phrases like "noVAE" which is incorrect, without a VAE absolutely zero image comes out. But bad nomenclature still spreads like wildfire. 🤷
noVAE means the default VAE attached to every checkpoint has not been overwritten with a finetuned one like the 840000-ema StabilityAI finetune. What this is really saying, is like "Engine Swapped" or "Not Engine Swapped".
And then as you said, if you run A1111/Vlad and have chosen a VAE in the options, then a model is loaded in full including the parameters for the VAE, which is then overwritten in place with your selected one every time (Engine Swapped).
@32Bitshifter thanks for explaination, still not clear to me what is better to use. i have 840000-ema wich i apply to basically every model in settings
I am very love your model, but can you upload a no VAE version? because I need to train some Lora based your model, VAE will result in very strange pictures, Appreciated!
Thank you! I'm doing it now, but I'm having trouble. As soon as I solve them, I will upload the noVAE model.
The noVAE version of the model has been uploaded.
@SG_161222 I've already tried to train a Lora model based it, and the results are amazing, thank you very much
What file shall I download, if I want the noVAE version?
Just found it =)
Details
Files
realisticVisionV60B1_v30VAE.safetensors
Mirrors
realisticVisionV60B1_v30VAE.safetensors
realisticVisionV30_v30VAE.safetensors
Realistic_Vision_V3.0.safetensors
RealisticVision.safetensors
Realistic_Vision_V3.0_fp16-no-ema.safetensors
realisticVisionV30_v30VAE.safetensors
realisticVisionV30_v30VAE.safetensors
realisticVisionV30_v30VAE.safetensors
realisticVisionV30_v30VAE.safetensors
realisticVisionV51_v30VAE.safetensors
realisticVisionV60B1_v30VAE_1.safetensors
realisticVisionV30_v30VAE.safetensors
RealisticVision.safetensors
realisticVisionV60B1_v30VAE.safetensors
Mirrors
realisticVisionV60B1_v30VAE.safetensors
Realistic_Vision_V3.0.safetensors
realisticVisionV30_v30VAE.safetensors
Qpipi.com_Realistic_Vision_V4.0.safetensors
Realistic_Vision_V3.0.safetensors
Realistic_Vision_V3.0.safetensors
realisticVisionV30_v30VAE.safetensors
realisticVisionV60B1_v30VAE.safetensors
Realistic_Vision_V3.0.safetensors
Available On (2 platforms)
Same model published on other platforms. May have additional downloads or version variants.








