Join The Tinkerer on Whop to get early releases, private pages, and the Tinkerer Discord role - all in one place. š Join on Whop
Semi-Real is honestly a pretty personal project. Itās changed and grown right along with everything else weāve done in CyberRealistic. At first, we kind of shelved it since there were already tons of semi-realistic models floating around. So it just sat in the lab for a while. Then Pony brought the idea back to life.
Now itās finally out in the wild. You get a super flexible, expressive SD1.5 model that cranks out awesome results, whether youāre making detailed portraits, trippy landscapes, or even going the NSFW route. Long prompts, short prompts, total chaos with wildcardsāit just handles whatever you throw at it and still looks great.
š Why Use It?
Total freedom with style (no stuffy rules about being ultra-realistic)
Works just as well for SFW or NSFW stuff
Super responsive, even if you only give it a couple of words (or go nuts with long prompts)
Comes with its own custom VAE, so colors and tones are dialed in
āļø Recommended Settings
Sampling method: Euler a or DPM++2M Karras
Clip skip: 2
VAE: custom made VAE is included but you can use your own VAE like vae-ft-mse-840000-ema-pruned or kl-f8-anime2
Sampling Steps: 16-40
HiRes Upscalers: 4x_foolhardy_Remacri / 2x_KemonoScale_v2
CFG Scale: 4-7
For the best results, it is recommended to use Adetailer (After Detailer) with the face detail model if the face does not look great.ā Negative Prompt Suggestions
cartoon, painting, illustration, (grayscale:1.4), (worst quality:2), (low quality:2), (normal quality:2), ugly, fat, 3D rendering, Asianless realistic
(worst quality, low quality:1.4)š¬ Prompt Assistant
Need better prompts for CyberRealistic Semi-Real? Try the custom ChatGPT prompt engine:
š [Generate the perfect prompt here]
ā Want to Support the Project?
If this model has helped you create something beautiful, wild, or just plain cool ā consider [buying me a coffee]. Every bit of support fuels the next generation.
ā ļø Friendly Reminder
This model can produce mature content. Please use it responsibly and ethically.
Description
In V3, several changes have been made, including the addition of CyberRealistic 7 & 8 training and the fixing of some bugs.
It remains extremely easy to achieve amazing results with this model. Still my favorite.
FAQ
Comments (16)
i never really understood the appeal of semi-realistic models
I think it can be particularly strong in creating fantasy images. Light effects are especially powerful in this model, and in realistic models, that aspect tends to fade a bit.
@Hamsome_Skidword Just curious, but how would you describe this style? It's not really realistic, but it's not illustrative either, right? (nice Lora's btw!)
https://civitai.com/models/1315394/gyaru-bitch-turned-gyaru-change-a-character-illustrious-pony
@CyberdeliaĀ that model isn't a style model, but the style is indeed semi-real. but i would rather go full illustrated, or full photoreal. its just that Pony Diffusion V6 has a strong semi-real base style, and that's what i used for the example images. i don't hate semi-real of course, but a lot of the semi-real AI images i see have terrible anatomy. though you seem to have done a really good job with this model, the quality is really high
@Hamsome_SkidwordĀ For me Semi-real always breaks the uncanny valley that jumps out at me in realistic images so I vastly prefer it.
@BrennokĀ I get that. Like Flux is more semi-real than real to me because of that. semi-real just hurts my eyes to look at because I have no idea how you could make images like that without AI, and I prefer my AI to not look like AI (which is why I don't like the style of Pony, or Flux)
@Hamsome_SkidwordĀ Lol, thatās true. Semi-realism isnāt a traditional 'art style' but it has existed long before AI - just look at Norman Rockwellās work. What AI has done is mass-produce a specific type of semi-realism, making it feel more like fast food: easy to consume, but not quite the same as traditionally food.
... but I really like fast food ;)
@CyberdeliaĀ that's true. surrealist art has been around long before i was born. I'm just not a fan of the AI version
At V3.0 you look like you are offering one of the best models on the site. But it is a SD1.5. I have given up on 1.5...
Update MAN! Please.
Update to what? Should I update to SDXL, Pony, or Illustrious? I might need to explore these options further....
@CyberdeliaĀ that's depends on you preferences mostly, because Pony and Illustrious (and NoobAI) SDXL based, so you can use sdxl loras with any of them. From my POV noob and illustrator better for semi-realism, pony are overtrained in pony/furry base and i really hope on v7 pony to fix score line in prompt
@RAI_34Ā check my profile ;) I have some... experience with Pony/Illustrious and SDXL
@CyberdeliaĀ You have some of the best checkpoints created. Just scrolling through the images in the gallery, it is hard to believe this is 1.5. Excellent work. For an update I am hoping that we get Reve Half Moon.
@wooba001466Ā not everything is 1.5. But I still use it the most. Keep the spirit of SD1.5 alive ;)
@CyberdeliaĀ a semi-real version of Illustrious or NoobAI from you would be epic
thanks for the update. v3 background detail and coherency is even better