GoR_v6ILXL+VAESETTINGS:DPM++ SDE Karras, DPM++ 2S a UniformEULER A Karrasclip skip 2Steps +/-30GFC+/-7Well, I committed this version. Take into consideration that trying to make this model realistic may have messed with the base illustrous a bit. The effects may be surprising ;)SDE can sometimes burn through but for that Euler A is finally useful.You can also test clipskip 1, but for me it is too “smooth”PS: And one more thing, if you see that I used your original prompt, please link the post in the comment under the image. Sometimes I'm surfing the civitai gallery and testing the prompts on my models and then just can't figure out what's from where<hr /><hr /><hr /><hr /><hr /><hr /><hr />-GoR_PONY_v5ilustros_H+VAE!This is experiment, it is Pony but with a liitle of Illustrious, lets say it is beta testSETTINGS:DPM++ SDE Exponentialclip skip 2CFG: 4,5 to 6,5Use negative prompt not only pony onesCheck my pics for settings<h3 id="please-upload-your-work-it-will-allow-me-to-improve-the-model.-i-myself-don't-have-much-time-lately-to-do-thorough-tests">Please upload your work, it will allow me to improve the model. I myself don't have much time lately to do thorough tests</h3><h3></h3><hr /><hr /><hr />-GODDESS of Realism - GoR_PONY_v4photo+VAE!bSampling method: DPM++ SDE KarrasSampling steps: 25 up+CFG Scale: 6.5Width: 832Height: 1216Clip skip 2For distance shot recomended ADetailer for better facesOthers:I don't quite know if it makes a difference but it seems to me that it improves the quality of the imagesClip skip SDXL onEmphasis mode No normEnable T5 <hr />The models used initially before my own training to create base GoR Pony were definitely Pony Realism and probably CyberRealistic Pony, DucHaiten-Pony-Real, Zonkey - Realism, DAMN! and in v1 Everclear PNY. I don't remember all exactly because I had probably all the realistic PONY models on disk. Thank you to all the creators.<hr />GoR pony v3The basic idea is to create the most versatile model, from nice views to NSFW and from robots and monsters to fury. V3 derives from v2art but managed to get crisp colors like in beta or 1. While the quality of textures is 100x better. I noticed that this version likes tags from sd1.5 such as HD, cinematic etc.... Also, the negative long shuttles have a very big impact but be careful.Beware of too big values especially for tags related to lighting (1.3), (1.5) and too much (((()))) etc.The model has a lot to be perfect, I'm still not satisfied with the face and hands in the distance shots. If anyone finds some lore or embedding that improves this please share in the comments.EnjoyV2art_fix+VAE!All DPM++ works good,for scifi/art 2m SDE and 2m SDE Heunfor human realism best DPM++ 2S a and DPM++ 3M SDE, DPM++ SDE KarrasCFG scale 8-9Steps 30upVAE baked inclipskip 1wuler and euler a not recomended<hr />Edit : The size of the model is a consequence of using the "checkpoint model mixer." I checked with other settings and it just comes out that way. As you can see from the added images it works correctlyEdit 2: LORAs trained on Pony Diffusion and Pony Realism works good. On Everclear not.I turned on "Lora/Networks: use old method" in a1111 settingsGoR_PONY_V2art:I added some custom lora to change mood of imagesIt is very sensitive model so don't go to high with ((()))) and (1.4)All DPM++ works good, I prefer 2M karrasGFC from 5 to 7Steps 30upVAE standard sdxlclipskip 1If you see others problems sher it so I can fix them in next versiontested in automatic1111GoR PONY_v1GFC works up to 10DPM++ 2M karras = fastDPM++ SDE karras = quality3MSDE, Euler, Euler a = works fine toSteps= 30-45VAE standard sdxlBe careful with adding too many tags like: photorealistic, photography, bokeh etc.... Sometimes their excess causes errorsGoR PONY_betaDPM++ 2M SDE karras-30steps for simple image and 40-50 for more complicated,for more info check posted images dataset and compare of all samplersGFC: 2 for realistic up to 7 to get more vibrantPony "scores" and tags for positive and negative SD:1.5 description belowThis is a mix of very many models from this site. The main goal was to create more artistic photographs in different styles.Prompt;photo type &gt; angle &gt; subject &gt;subject details &gt; background &gt; quality tags (not necessarily) &gt; style (not necessarily)samplersbest with DPM++ SDE Karrassteps 25+ gfc 7hires fix as you like ( I prefer foolhardy)Clip Skip 1It dislikes "full body shot" use full shot and othersNegative promptdangerdonkey, kkw-new-neg-v1.6 ,(low water quality:1.45), low particles efects, ( 3d render, anime, draw, cartoon, grotesque),( over saturated:1.1), jpeg artifacts, out of frame, (((watermark, signature))), lowres, (low quality skin:1.1), , error, compressed, 000, dolly skin, (plastic doll), barbie, (lolicon, loli:1.2), childlike, 000, bad anatomy, bad hands, mutated hand, missing fingers, fused fingers, extra fingers, extra arms, long neck, amputation, (poor quality feet:1.0), wrong number of feet, blending objects, extra digit, fewer digits, cropped, out of frame, error, extra limbs, mutation, bad proportions, cropped head, malformed limbs, mutated hands, mutated feet, missing limbs, cropped, deformed, ugly, mutilated, disfigured, face cut, head cut, legs cut, poorly drawn face, animals, nsfw,

GODDESS of Realism

New marging method, new tool. More time because from 1.5 to even 5 hours to combine two models. But I think it was worth it. What checkpoints were used I do not remember, at the best of times I had about 150 on disk.STEPS 30-40GFC 5-7HIRESS as you likeEnabled afterdetailer if you want better facesJust go through the data from the images and you will see what and how

Files with SHA256: 99e314ce754b751dd69acaec7a7df56b193392ceaa2be610271c84366ed58f43