We have updated our quality standards found under our uploading guidelines.
This was done mostly because modern AI generations are of much better quality now than they have been in the past, and we would like for this archive to focus on displaying the best work, not just any and all work. Please take a moment to review the guidelines before uploading going forward now. Older submissions will be exempt for this at this point in time, but keep in mind in the far future we might clean those up as quality continues to improve.
parszuki
MemberHowl with joy!!!
Beautiful long hooves, a lighter patch of fur around the rump and hind hooves, and that raised tail. Plus a slightly rounded rump... Yum! Sambaras have a habit of raising their tail like a flag when they are disturbed. They also ruffle their tails intensely. The only thing that cannot be captured in the graphics is the sound that an anxious sambar makes - it resembles a car horn :)
https://www.youtube.com/watch?v=RpwozInmXG0
Was this artwork the starting image for your work?
https://www.flickr.com/photos/mshehan/7037622281/sizes/o/
Lungfish1223
MemberNot this one in particular, though that'd probably be a very easy one to adapt as well. Most of my references I get from iNaturalist since you can easily see the license the photo is under, though flickr offers a similar ease (I am not 100% sure on the legality of using "all rights reserved" as base images, but I'd like to avoid doing so regardless, just as a precaution) and, well, with most species iNat has a wide range of data available.
parszuki
MemberThis sambar itself looks great. Super fluffy! I see the information: License: Some rights reser.
Click on the details and check whether you can use it.
Sambar would definitely be happy with the improvement! ^_^
parszuki
MemberI found your image on discord and clicked the "magnifying glass" icon. I was surprised to receive the prompt and settings! You used the img2img method! It was the same with your beautiful kudus.
Prompt
ass focus of a feral deer standing in a forest and raising her tail teasingly, (excessive genital fluids, drooling), (masterpiece, detailed background, high detail, absurd res), (by claweddrip), (by dimwitdog:teranen:0.35), by personalami, (feral female:1.2), tongue out, (detailed fur, fluffy, detailed eyes), (visible pussy:1.5), (big pussy:1.3), (pussy juice:1.2), in heat, thick thighs, bedroom eyes, twitching, breath, (mature female:1.3), big anus, brown fur, white chin, light brown underbelly, dark brown eyes, <lora:more_details: 0.45>, <lora:fluffyrock-quality-tags-v3.0-vpred:1>
Negative Prompt
boring_e621_v4, FastNegativeV2, deformityv6, bwu, dfc, ubbp, updn, worst quality, low quality, (anthro:1.3), (text, logo, watermark:1.5), (human hair:1.3), (dyed hair:1.3), (spotted fur, antlers:1.3), (buildings:1.5)
Steps: 30
Sampler: DDIM
CFG scale: 7
Seed: 875414406
Size: 2048x2048
Model hash: 1ac4dcb22c
Model: EasyFluffV10.1
Denoising strength: 0
CFG Rescale: 0.7
fluffyrock-quality-tags-v3.0-vpred
27c5fb166e9e"
Version: v1.6.0-2-g4afaaf8a
Unfortunately, the use of "EasyFluff" ended in failure (a deer's ass in black pixels) and the use of Yifymix or bb95 resulted in poor realistic graphics with numerous errors.
Where did Discord get the prompt data - did it read from tea leaves?
Lungfish1223
MemberEasyFluff is a vpred model and so needs some accompanying configuration files and CFG rescale installed. I am not sure civitAI supports that. Without these, EasyFluff won't be able to do much of anything - it'll just fall apart. For models that are not vpred, however, my prompts will likely not get desired results - vpred models can function with natural language, others not so much.
I believe one of the discord bots automatically reads the metadata of images posted. All the images I upload, both on discord and here, have intact metadata so websites and programs able to read it should be able to show the prompts.
On a sidenote, the metadata shows denoising strength 0 - that's a lie. The last run was at denoise 0 to put the metadata back in its place, but the actual denoise level the image was generated on was somewhere around 0.35-0.45
Updated
parszuki
MemberApparently there are some hybrid methods - you have to download the files and embed them in the appropriate location, then change something in the browser (in the advanced page settings?) and apparently it works. But it's black magic for me at the moment :(
And I think I have an "indefinite leave from generating":
https://civitai.com/articles/3307
+and now the website currently has an ultra-error:/
Login to respond »