Model
Trends
.ai
Model
Trends
.ai
the best open source ai models
Run Locally
shintaisou-il.safetensors
218 MB · SafeTensor
Downloads are served by Civitai. Some creators ask for a Civitai login first.
AD
Install / Run on Graydient
Requires membership
<shintaisou-xl>
About This Model
STABLE DIFFUSION XL 1.0 · LoRAs
Created by
karetai
on Civitai
by karetai - Shintaisou (Shin) - Princess 69: Midnight Gymnastics
Creator notes
Dark Rhythmic Gymnastics I never played the visual novel, but I am familiar with the OVA, even though it felt like the budget for the second episode had halved. This is trained on most of the game images although I only took a sample of character portraits because there's a billion of them. It gets things mostly right, although pairing (and larger groups) characters may take a bit more effort and or lucky rolls of the dice. It's not my favourite series, and there's some pretty weirdly fetishistic imagery that isn't really my jam (but you do you). That said I think this Lora turned out okay. Also whoever decided it should be called Princess 69 in English, just, why. V3: In some ways better than V2, and in some ways worse. I tried some different approaches to get multi-character in the one scene to be more consistent and... the results are mixed. I'm not super happy with it, but here you go. I'm going to leave this one for now, unless I have some breakthrough on how to make it reliably better. V2: V2 was a real headache. I had to do a lot of gymnastics (ha) to get it to behave a little more predictably. It still tries to create three people instead of two when in landscape, maybe I have to increase the base training size to 1280. Anyway, it's a little more predictable overall than V1, and is the longest training I've done. Some pairings of characters work better than others. Character names are now all one word as I found when it was two words, it was less predictable, same with underscores which I think get converted to spaces anyway. Matching clothing serafuku and leotard are the two main types of clothing. I've changed the leotards like so: pink and beige leotard - Fubuki, Koyuki green dress, turtleneck, layered clothing - Miku dark blue leotard - Mizuki purple and white leotard - Tomomi (a little less reliable, try purple leotard too) orange and brown leotard - Ikuko (also less reliable, try orange leotard too) blue and white leotard - Lu Other note, if you're putting two girls together, you may also find at least specifying the hair length useful, such as shirakawatomomi, fujimiyakoyuki, 2girls, long hair, short hair and it'll generally get it more than just the character triggers alone. V1: Matching clothing Generally dark grey serafuku, white neckerchief will get you the school uniform. Though some characters share leotards, the correct combinations are: Fubuki, Koyuki: pink leotard Miku: green dress, turtleneck (or you can try for the single time she has awhite leotard). Incidentally she won't be showing unless you prompt her so. Poor Miku. Mizuki: dark blue leotard Tomomi: purple leotard and sometimes black dress, black bow ribbons. Getting Fubuki to wear the black dress in the same picture as Tomomi is a challenge for the reader. Ikuko: orange leotard (questionable if it's orange but that's what we're going with) Lu: light blue leotard Don't ask me why Civitai classified anime-style characters in leotards as R, yet Miku is PG - seems a bit excessive. Must be the occasional extra/long finger, body horror n all that. This is also my first time training against Illustrious 2.0 (instead of 0.1)
Trigger words
shirakawatomomi, 1girl, solo
takemotoikuko, 1girl, solo
fujimiyakoyuki, 1girl, solo
tianlu, 1girl, solo
nonoharamizuki, 1girl, solo
haneokafubuki, 1boy, solo
haneokamiku, 1girl, solo
236 downloads · 41 likes on Civitai
SHA256
C586F059E1FF0DE70BB57A3ADB4950AD25D0582A9221566070ED406D79FEF4FD
ModelTrends.ai Model ID
#21448
Flag content
Example renders
Prompts and settings shared by the people who made these renders on Civitai. Pick a render to see how it was made.
Settings
Sampler
DDPM
Steps
20
Guidance
5
Seed
3365716763
Size
1024x1440
Prompt
<lora:shintaisou-il:0.9> shirakawatomomi, 1girl, solo, serafuku,, best quality, high quality, masterpiece,
Negative prompt
(worst quality, low quality:1.2), normal quality, lowres, greyscale, mosaic censoring, censored, signature, watermark, artist name, patreon logo, patreon url, duplicate characters,
Settings
Sampler
DDPM
Steps
20
Guidance
5
Seed
4273968794
Size
1024x1440
Prompt
<lora:shintaisou-il:0.9> haneokafubuki, shirakawatomomi, 1boy, 1girl, duo, hetero, serafuku, sitting, hug from behind, holding hands, assertive female, smug, concerned, skirt,, best quality, high quality, masterpiece,
Negative prompt
(worst quality, low quality:1.2), normal quality, lowres, greyscale, mosaic censoring, censored, signature, watermark, artist name, patreon logo, patreon url, duplicate characters,
Settings
Sampler
DDPM
Steps
20
Guidance
5
Seed
2925256754
Size
1024x1440
Prompt
<lora:shintaisou-il:0.9> nonoharamizuki, 1girl, solo, serafuku,, best quality, high quality, masterpiece,
Negative prompt
(worst quality, low quality:1.2), normal quality, lowres, greyscale, mosaic censoring, censored, signature, watermark, artist name, patreon logo, patreon url, duplicate characters,
Settings
Sampler
DDPM
Steps
20
Guidance
5
Seed
2816829473
Size
1024x1440
Prompt
<lora:shintaisou-il:0.9> haneokafubuki, 1boy, solo, serafuku,, best quality, high quality, masterpiece,
Negative prompt
(worst quality, low quality:1.2), normal quality, lowres, greyscale, mosaic censoring, censored, signature, watermark, artist name, patreon logo, patreon url, duplicate characters,
Settings
Sampler
DDPM
Steps
20
Guidance
5
Seed
4151667370
Size
1024x1440
Prompt
<lora:shintaisou-il:0.9> takemotoikuko, 1girl, solo, serafuku,, best quality, high quality, masterpiece,
Negative prompt
(worst quality, low quality:1.2), normal quality, lowres, greyscale, mosaic censoring, censored, signature, watermark, artist name, patreon logo, patreon url, duplicate characters,
Settings
Sampler
DDPM
Steps
20
Guidance
5
Seed
1523451302
Size
1024x1440
Prompt
<lora:shintaisou-il:0.9> haneokamiku, 1girl, solo,, best quality, high quality, masterpiece,
Negative prompt
(worst quality, low quality:1.2), normal quality, lowres, greyscale, mosaic censoring, censored, signature, watermark, artist name, patreon logo, patreon url, duplicate characters,
Settings
Sampler
DDPM
Steps
20
Guidance
5
Seed
3948750466
Size
1024x1440
Prompt
<lora:shintaisou-il:0.9> tianlu, 1girl, solo, serafuku,, best quality, high quality, masterpiece,
Negative prompt
(worst quality, low quality:1.2), normal quality, lowres, greyscale, mosaic censoring, censored, signature, watermark, artist name, patreon logo, patreon url, duplicate characters,
Settings
Sampler
DDPM
Steps
20
Guidance
5
Seed
1079688504
Size
1024x1440
Prompt
<lora:shintaisou-il:0.9> fujimiyakoyuki, 1girl, solo, serafuku,, best quality, high quality, masterpiece,
Negative prompt
(worst quality, low quality:1.2), normal quality, lowres, greyscale, mosaic censoring, censored, signature, watermark, artist name, patreon logo, patreon url, duplicate characters,
AI Models for Image, Video & Text Generation
shintaisou xl is a LoRA for the Stable Diffusion XL 1.0 family listed on ModelTrends.ai, a read-only catalog of open source AI image, video and text models. Compare it with other Stable Diffusion XL 1.0 models, check its heat score to see how it is trending, and open it on PirateDiffusion or BitVector to try it.
Browse by Model Family
Browse all models