Model
Trends
.ai
Model
Trends
.ai
the best open source ai models
Run Locally
GachaSplash4.safetensors
144 MB · SafeTensor
Downloads are served by Civitai. Some creators ask for a Civitai login first.
AD
Install / Run on Graydient
Requires membership
<gachasplash>
About This Model
STABLE DIFFUSION 1.5 · LoRAs
Created by
silvermoong
on Civitai
GachaSplash is a LoRa that will give our renders an interesting background effect. Unfortunately, it’s not entirely compatible. It still works and can produce good-looking results, but despite following the template created by the author, obtaining the desired background effect seems random. Template: [(white background:1.5)::5], isometric OR hexagon, 1 girl, mid shot, full body, Try it out with ,, or models. Each of them produces amazing results. Add wood and stone as negative prompts if you see too many of those elements. This can also be applied to certain clothing elements if the LoRa keeps producing characters with them. Suggested resolution: 640X640, then use /highdef on the best picture. In this case, treat the triggers more like suggestions of prompts to try out.
Creator notes
The training dataset is 100% generated by NovelAI. No real artwork is used. Checkout my Alter version, which uses real artwork: 3.0 version is released. I rebuilt the entire training set for this version. Notes: - Use hires fix + SD upscale or MultiDiffusion to get better face details. - Try adding wood, stone in negative prompt if you see too many woods, stones, structures. You may also emphasize it. For example, the bottle keeps getting transformed into structures till I add wood, stone. A 2.0 version is released: A gacha splash style lora. Make sure you start with the following template and add your background prompts. [(white Suggested resolution: 640X640 with hires fix. You may modify your prompt and resolution after successfully getting the effect. I trained two different versions with different learning rate and scheduler. Gacha Splash Fantasy fits more in abstract backgrounds. Gacha Splash is intentionally trained to be slightly overfit. It fits greatly for architectures. See compares from sample images Notes: 1. If your characters are always wearing jackets/half off jackets, try adding "off shoulder" in negative prompt. You may further add "jackets"/ "bare shoulders" if the issue persists. 2. Similar as the jacket issue, if you find any elements that are always being generated, try adding them into negative prompt first. The training dataset is well tagged. Therefore AI should be able to remove these elements. Prompts like hexagon and isometric help to shape the gacha splash effect. There are some alternative choices.
Trigger words
(white background:1.5)
[(white background:1.5),::5]
49,660 downloads · 5,956 likes on Civitai
SHA256
EEBDB8E46FAFDAF3FCA98D0B3ADB0CBB16F71E138B65DDF8E43288D0B4A511EB
ModelTrends.ai Model ID
#5710
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
Euler a
Steps
25
Guidance
7
Seed
3656898866
Size
832x1280
Clip skip
2
Prompt
score_9, score_8_up, score_7_up, score_6_up, 1girl, (sculpture art stylized by Jean-Baptiste Carpeaux:1.3) , dark graffiti style, fairy tale, Homely Dull ([Canyonlands National Park|The Swiss Alps]:1.3) in the distance there is a Bandar Seri Begawan, [Edo Period|Somali] foliage and Capricornus constellation in background, at Twilight, Maximalist, Disgusting, Fauvism, Light and shadow plays, Fish-eye Lens, anaglyph effect, Iridescent and fluorescent flakes, Simplified style, Unsplash, artstation, intricate, close-up <lora:add-detail-xl:0.8>, <lora:quality1:0.8>, <lora:aesthetic_anime_v1s:0.8>, <lora:detailed Dark dire prophecy landscape:0.8>, <lora:lihui4JXK-b2-bf16-128-128-1-re1-ep3-768-DA-5015fix:1> (white background:1.5), [(white background:1.5),::5]
Negative prompt
score_5, score_4, blurry, ugly face, bad anatomy, mutated hands, low res, blurry face, bad hands, wrong hand, bad fingers, wrong fingers, extra fingers, negativeXL_D,
Settings
Sampler
DPM++ 2M Karras
Steps
28
Guidance
7
Seed
4124061440
Size
512x768
Prompt
[(white background:1.5),::5] (isometric:0.6), a cute white ghost riding a bicycle in full speed down a street, Photography <lora:GachaSplash4:1>
Negative prompt
girl, human, girl, woman, women, (worst quality:2), (low quality:2), nsfw, nipples, text, watermark, (deformed, distorted, disfigured:1.3), poorly drawn, bad anatomy, wrong anatomy, extra limb, missing limb, floating limbs, (mutated hands and fingers:1.4), disconnected limbs, mutation, mutated, ugly, disgusting, blurry, amputation, three hands, three foots
Settings
Sampler
DPM++ 2M Karras
Steps
42
Guidance
11
Seed
3015181284
Size
640x512
Clip skip
2
Prompt
[(white background:1.5),::5] 1girl, mid shot, full body, ocean, sea waves, water splashes, sky, light particles, butterflies, night, starry sky <lora:lihui4JXK-b2-bf16-128-128-1-re1-ep3-768-DA-5015fix:1>
Negative prompt
bad anatomy, extra fingers, watermark, (worst quality, low quality:1.4)
Settings
Sampler
DPM++ 2M Karras
Steps
42
Guidance
11
Seed
795901665
Size
640x512
Clip skip
2
Prompt
[(white background:1.5),::5] (isometric:1.0), double exposure, bubble, mid shot, full body, // masterpiece, best quality, 1girl, (colorful),(finely detailed beautiful eyes and detailed face),cinematic lighting,bust shot,extremely detailed CG unity 8k wallpaper,black hair,long hair,black eyes,solo,smile,intricate skirt,((flying petal)),(Flowery meadow) sky, cloudy_sky, building, moonlight, moon, night, (dark theme:1.3), light, fantasy,looking at viewer, <lora:lihui4JXK-b2-bf16-128-128-1-re3-ep1-768-DA-5015fix:0.85>
Negative prompt
bad anatomy, extra fingers, watermark, (worst quality, low quality:1.4)
Settings
Sampler
Heun
Steps
30
Guidance
15
Seed
1648914113
Size
704x448
Clip skip
2
Prompt
<lora:genshinfull1:1>,best quality,monochrome,<lora:lihui4JXK-b2-bf16-128-128-1-re1-ep3-768-DA-5015fix:1>,1girl,solo,niloudef,water spash,gentle smile,outline,<lora:animeoutlineV4_16:1>,genshin impact,lineart,outline,red_hair,
Negative prompt
badhandv4,ng_deepnegative_v1_75t,(worst quality:2),(normal quality:2),(low quality:2),low res,blurry,text,watermark,logo,banner,extra digits,cropped,jpeg artifacts,signature,username,error,sketch,duplicate,ugly,monochrome,horror,geometry,mutation,disgusting,bad anatomy,EasyNegative,BadNegAnatomyV1-neg,FastNegativeV2,
Settings
Size
1024x1536
Prompt
(masterpiece:1.1), (ultra-detailed), (best quality), macro, 8 K, cinematic lighting, A close-up macro scene of a swallowtail butterfly resting on a rugged sandstone cliff at golden hour. Its wings are half-open and shimmer like a galaxy — sapphire blue, teal, magenta, and amber streaks, dusted with fine glittering specks. Bold black veins trace the wings; the lower edge shows vivid orange-red teardrop spots. Back-lit by warm sunset light that pours through forest haze, forming soft volumetric rays and a golden rim-light along every wing edge. Tiny glowing butterflies drift in the bokeh background; floating dust motes catch the light. Photorealistic wing-scale texture, crisp focus on the butterfly and rock grain, dreamy blurred backdrop, fantasy yet natural atmosphere.
Negative prompt
(worst quality:1.2), low quality, lowres, blurry, pixelated, jpeg artifacts, text, watermark, signature, cartoon style, oversaturated, dull lighting, flat shadows, overexposed, harsh contrast, muted colors, disfigured anatomy, mutated wings, extra limbs, cutoff parts, duplicate subjects
Settings
Sampler
Heun
Steps
31
Guidance
12
Seed
126872127
Size
704x448
Clip skip
2
Prompt
<lora:genshinfull1:1>,(luminedef),gentle smile,best quality,(lineart:1.5),(outline:1.5),monochrome,<lora:animeoutlineV4_16:1>,<lora:lihui4JXK-b2-bf16-128-128-1-re1-ep3-768-DA-5015fix:1>,1girl,solo,outline,blonde hair,rainy days,
Negative prompt
badhandv4,ng_deepnegative_v1_75t,(worst quality:2),(normal quality:2),(low quality:2),low res,blurry,text,watermark,logo,banner,extra digits,cropped,jpeg artifacts,signature,username,error,sketch,duplicate,ugly,horror,geometry,mutation,disgusting,bad anatomy,
Settings
Sampler
DPM++ 2M Karras
Steps
42
Guidance
11
Seed
2822546827
Size
640x512
Clip skip
2
Prompt
[(sky background:1.5),::5] (isometric:1.0), double exposure, bubble, mid shot, full body, // [(white background:1.4),::10], // book, bookshelf, brown_hair, open_book, braid, solo, 1girl, long_hair, blue_eyes, collarbone, long_sleeves, book_stack, chair <lora:adaptedmodel:1> <lora:lihui4JXK-b2-bf16-128-128-1-re1-ep3-768-DA-5015fix:1.0>
Negative prompt
Weird image, bad anatomy, extra fingers, watermark, (worst quality, low quality:1.4)
AI Models for Image, Video & Text Generation
gachasplash is a LoRA for the Stable Diffusion 1.5 family listed on ModelTrends.ai, a read-only catalog of open source AI image, video and text models. Compare it with other Stable Diffusion 1.5 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