Model
Trends
.ai
Model
Trends
.ai
the best open source ai models
Run Locally
velvetteXLFurryAnimeMore_v4NoblesseFOCUS_2314715.safetensors
6.62 GB · SafeTensor · fp16 · full
velvetteXLFurryAnimeMore_v4NoblesseFOCUS_3062548.safetensors
6.46 GB · SafeTensor · fp16
Download
velvetteXLFurryAnimeMore_v4NoblesseFOCUS_3062899.safetensors
6.62 GB · SafeTensor · fp16
Download
Downloads are served by Civitai. Some creators ask for a Civitai login first.
AD
Install / Run on Graydient
Requires membership
<velvette4-xl>
About This Model
STABLE DIFFUSION XL 1.0 · Base Models
Created by
Herrscher_AGGA
on Civitai
Creator notes
👋 If you like what I do and want to support the development, feel free to buy me a coffee: Velvette-XL ✨ Velvette-XL is the reconstruction of an unreleased experimental model—recovered, refined, and expanded into a flexible hybrid checkpoint. While earlier versions captured a very distinct identity, version 4 moves in a broader direction: it carries more visual information, more styles, and much wider compatibility. Built from a fusion of diverse 2D styles—from expressive sketch lines to polished semi-realistic rendering—Velvette-XL focuses on line versatility, adaptable detail, and creative control. It can shift between defined lineart, lineless illustration, and highly detailed compositions depending on prompt weight and CFG range. Originating from the Providence concept, Velvette-XL preserves the multi-style, lightweight philosophy of its earliest versions, now expanded into a more complete, more capable, and more flexible model—one that rewards intentional prompting while still retaining the roots of its original identity. V4 | Noblesse FOCUS I've been using sd-mecha since v2, and by the time I reached v4 I ended up working with way too many versions (I even thought about giving up at one point, because it was generating nothing but noise). I was so used to SuperMerge that I had to completely rethink my entire merging workflow. I switched back and forth between add_difference and weighted_sum until I finally adapted to this new method—which turned out to be incredibly smooth and fast. Version 4 went through several full restarts (the first successful compilation is also merged here… almost as I remember it), and there are more than twenty internal iterations (5–10 models each). Some were better, some were worse, but the goal was to gather as much information as possible and then improve visual quality—although it wasn’t really necessary, since iterations 18–19 were already perfect. In short, this version contains a large number of merges. It brings together furry, hentai, 3D, 2D, anime, and more. Velvette v4 is a model designed to preserve the maximum amount of information possible. (Though I still need to add some extra pixel-level data—but I forgot.) Better Quality Prompt: Positive: (masterpiece, high quality, best quality, amazing quality, very-aesthetic, newest, scenery, very awa):8, (detailed, ultra-detailed, addmicrodetails, intricate details):8, (8k, Ultra HD, high resolution, absurdres, incredibly absurdres, hdr):8, (ultra detailed face, detailed eyes, extremely detailed eyes, perfect eyes, detailed iris):8, (thick lines, bold outlines, inked lines, strong contour, cel shading, digital illustration, stylized shading, soft brushwork, painterly edges):8, Negative: (worst aesthetic, flat color):8, (bad quality, low detail, worst quality):8, (bad eyes, bad iris, undetailed eyes, undetaled face, bad face):8, (bad anatomy, bad hands, extra fingers, distorted hands):8, (malformed, deformed, distorted, mutated, blurry):8, (patreon username, twitter username, artist name, signature, web address, weibo watermark, watermark, url):8, Style Keywords (positive or negative): (realistic, photorealistic, hyperrealistic, 3d, 2.5d, cgi, anime, cartoon) V3 | ALPHA ALPHA is an incomplete version—a fusion of v1 + v2 intended to be merged later with IL v2.0 or v3.0, as well as rouwei v7. I’ve tested it for days and ended up liking it much more than expected, so I didn’t finish its final merge. There is also a sister version including Pony Diffusion, but I haven’t tested it enough yet. (Side note: I was a bit disappointed with Civitai, but I managed to merge checkpoints on free Colab accounts—my wallet is thankful.) V2 | EPS There are changes! For V2, I tried to beef up Velvette, but the same thing happened as in v1.1 (it significantly increased the strength of the prompts), so I reduced the strength of the base style (something I wanted to do), doubled the strength of PVDC, and merged the v1.1 excerpt to 0.1 strength. This improved the model, and I simply merged it with noobai eps (I was told it was good for reducing steps, although I don't know what changed because v1 had dmd2 and I added it this time). In short... I don't know what differentiates this from that, and I've only tested it with Illustrious lora, so I don't know how much compatibility it has. Anyway, I like the result. By the way, it also works with the VVTT_detailer. Thanks for reading, bye~ 7w7)/ 🎛️ Behavior with CFG and Steps - CFG 1–4 → Faithful to applied LoRA, cleaner lines. - CFG 4–7 → Velvette’s base style takes over, sketch-like look. - CFG 7–8+ → More saturated and cartoonish appearance, similar to using a style LoRA at strength 1.5–2. 📌 Practical tip: If you’re not using LoRAs at low CFG, add: - Positive → hdr:2.5 - Negative → flat This adds extra detail even at CFG 1–2, reaching quality comparable to Utopia v4 at CFG 4. 🔹 Step Performance - Minimum useful → 6 steps at CFG 1. - Optimal → 12–30 steps. ⚙️ Quick Guide - Steps → 10–30 (optimal: 12–30) - CFG → 1.5 for smoothness | 3+ for stronger definition - Samplers → DPM++ 2M Karras, Euler a, turbo, etc. - Resolution → 1024×1024+ (1152×1344 and 1536×896 work very well) - Upscalers → R-ESRGAN 4x+ Anime6B / 4x-UltraSharp 🏷️ Useful Tags Basic styles: pvdc_il, pvdc_stl, sketch style, pastel style, cartoon style, thick lines, hatched style, comic style, detailed style, fabulosity style, cutecore, horrorcore, glitchcore, cutelatte, anime, shiny Examples: - (sketch style + thick lines + hatched style) - (comic style + detailed style) - (pastel style + fabulosity style) 🎨 Prompting Tips General positives: (masterpiece), (best quality), (ultra-detailed), perfect composition, intricate details, official wallpaper By theme: - Anime/Cartoon → anime, cutecore, shiny skin - Realistic → (REALISTIC:1.15), detailed eyes, realistic lighting - Experimental → glitchcore, watercolor, abstract, dark theme Recommended negatives: bad quality, wors
2,206 downloads · 544 likes on Civitai
SHA256
630D759B066312B3A25685FD8A94CAF1A8C459EBCF445260A09412D6505CCE18
ModelTrends.ai Model ID
#19225
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
14
Guidance
6.5
Seed
412910
Size
1024x1536
Clip skip
2
Prompt
(digital illustration, stylized shading, soft brushwork, painterly edges):2, (masterpiece, best quality, ultra detailed, 8k), dynamic portrait angle, 1girl, mafia woman, cigarette, blood on face, intense gaze, upper body, stylish composition, simple white background, illustration, depth of field
Negative prompt
(photorealistic, hyperrealistic, cel-shading, anime-style:1.2), (bad quality, low detail:1.3), (deformed, distorted, mutated:1.1)
Settings
Sampler
DPM++ 2M
Steps
14
Guidance
5
Seed
2200754347
Size
832x1216
Clip skip
2
Prompt
(masterpiece, masterwork, best quality, amazing quality, very aesthetic, ultra-detailed, high detail, detailed, absurdres, superabsurd res, high resolution, hi res, HDR, 8K, 4K, Ultra HD, newest, beautiful, beautiful face, cute girl, very awa, absolutely eye-catching, extremely detailed eyes, EyesHD:1.2, clear skin, eyelashes, sharp focus, depth of field, high contrast, brightly lit, highly visible, perfect lighting, best lighting, volumetric lighting, light rays, ray tracing, raytracing, ambient occlusion, soft lighting, cinematic, cinematic composition, epic, detailed background, intricate cinematic background, detailed illustration, scenery, illustrating, CG, painting_(artwork), painting \(medium\), oil painting \(medium\), pastel \(medium\), hatching \(texture\), fabulosity style, chiaroscuro, pvdc_il:3.5, PVDC_STL:3.5, hatched style:3.5, sketch style:3.5, thick lines:3.5, (((pvdc_il, pvdc_stl, hatched style, thick lines)))), colorful ink splash, intense expression, tenebrism, blurry foreground, cool theme, (close-up:1.5), poster \(medium\), artbook, dark maniac theme, textured painterly, BREAK tamamo \(fate\), 1girl, solo, intricate details, masterpiece, best quality, absurdres, detailed face, detailed eyes, skindentation, blurry background, , , anime coloring, sheer nightgown, private suite, shirt slightly open, , dreamy , hugging knees, legs apart, half-lidded eyes, slightly flushed skin, depth of field, close-up, straight-on view, BREAK fizrotart:0.5, 1=2:0.65, gloss \(hotglossed\):0.25, (by rintaro komori:0.53), (by ilya kuvshinov:0.5), (by aoi ogata:0.5), (rella:0.42), (au \(d elete\):0.12), (dino \(dinoartforame\):0.31),wlop:0.12, (qiandaiyiyu:0.45),(soleil \(soleilmtfbwy03\):0.6),(anniechromes:0.5),
Negative prompt
low quality,bad quality,worst quality,worst detail, lowres, (low quality:1.2), censored, mosaic censoring, anatomical nonsense, bad anatomy, worst aesthetic, old, mismatched pupils, extra limbs, bad hands, deformed hands, missing finger, bad fingers, watermark, artist name, signature, logo, patreon logo, patreon username, web address, email address,
Settings
Sampler
DPM++ 2M
Steps
18
Guidance
5
Seed
1713404134
Size
832x1216
Prompt
masterpiece, best quality, very aesthetic, high resolution, ultra-detailed, absurdres, newest, extremely detailed eyes, beautiful, perfect lighting, hi res, amazing quality, very awa, Brightly Lit, Highly Visible, 8k, sharp focus, beautiful, perfect lighting, volumetric lighting, scenery, painting_(artwork), 1girl, chen, touhou, brown hair, brown eyes, cat ears, green mob cap, single earring, fang out, peeking out, climbing, two tails, nekomata, (matcha:0.95), (ciloranko:0.95), (modare:0.95), (Cutesexyrobutts:0.95), (WLOP:0.95), (guweiz:0.95), dark theme, night, ilya_kuvshinov, amonitto, kankan33333, dino \(dinoartforame\), modare, Quasarcake, mika_pikazo, close-up, face focus, eyelashes, outdoors, nature, tree, blurry background, depth of field, bokeh, (lens flare:1.1), amazing quality, very aesthetic, absurdres hdr:0.885 (pvdc_il:2.63, pvdc_stl:2.63)
Negative prompt
(worst quality:1.2), (low quality:1.2), (normal quality:1.2), lowres, bad anatomy, bad hands, signature, username, logo, artist name, watermark, patreon username, bad details,
Settings
Sampler
Euler a
Steps
14
Guidance
6.2
Seed
21733102
Size
832x1216
Prompt
masterpiece, best quality, absurdres, newest, realistic, photorealistic, 3d, pinup \(style\), 1girl, solo, jessie \(pokemon\), pokemon \(classic anjme\), mature female, red hair, long hair, widow's peak, looking at viewer, blue eyes, makeup, lipstick, earrings, sphere earrings, nose, lips, smile, smirk, smug, v-shaped eyebrows, furrowed brow, breasts, large breasts, team rocket, team rocket uniform, white shirt, skinny, petite, eye focus, portrait, close-up, perspective, dutch angle, depth of field, white background, simple background hdr:1.885 (pvdc_il:2.63, pvdc_stl:2.63)
Negative prompt
low quality:4.5,bad quality:4.5,worst quality:4.5,worst detail:4.5, censor, patreon username, patreon logo, watermark:4.5, artist name:4.5, signature:4.4, logo:4, anatomical nonsense, bad anatomy, worst aesthetic, old, mismatched pupils, extra limbs, bad hands, bad fingers, text:4.5, web address, email address,
Settings
Sampler
DPM++ 2M
Steps
14
Guidance
5
Seed
3875901267
Size
832x1216
Prompt
(masterpiece, masterwork, best quality, amazing quality, very aesthetic, ultra-detailed, high detail, detailed, absurdres, superabsurd res, high resolution, hi res, HDR, 8K, 4K, Ultra HD, newest, beautiful, beautiful face, cute girl, very awa, absolutely eye-catching, extremely detailed eyes, EyesHD:1.2, clear skin, eyelashes, sharp focus, depth of field, high contrast, brightly lit, highly visible, perfect lighting, best lighting, volumetric lighting, light rays, ray tracing, raytracing, ambient occlusion, soft lighting, cinematic, cinematic composition, epic, detailed background, intricate cinematic background, detailed illustration, scenery, illustrating, CG, painting_(artwork), painting \(medium\), oil painting \(medium\), pastel \(medium\), hatching \(texture\), fabulosity style, chiaroscuro, pvdc_il:3.5, PVDC_STL:3.5, hatched style:3.5, sketch style:3.5, thick lines:3.5, (((pvdc_il, pvdc_stl, hatched style, thick lines)))), colorful ink splash, intense expression, black and white theme,tenebrism,blurry foreground,split theme BREAK solo,1girl,fern \(sousou_no_frieren\),long_hair,black_hair,purple_eyes, looking at viewer,nail polish,portrait,nails,eye focus,finger frame,head_tilt, BREAK (qiandaiyiyu:0.85),(soleil \(soleilmtfbwy03\):0.6),(godiva ghoul:0.65),(anniechromes:0.5), (rella:1.2),(redum4:1.2) (au \(d elete\):1.2) (dino \(dinoartforame\):1.1),wlop, red eyes
Negative prompt
low quality,bad quality,worst quality,worst detail, lowres, (low quality:1.2), censored, mosaic censoring, anatomical nonsense, bad anatomy, worst aesthetic, old, mismatched pupils, extra limbs, bad hands, deformed hands, missing finger, bad fingers, watermark, artist name, signature, logo, patreon logo, patreon username, web address, email address,
Settings
Sampler
Euler a
Steps
14
Guidance
6.2
Seed
4189778181
Size
1094x1606
Clip skip
2
Prompt
(pastel \(medium\), oil painting \(medium\), painting \(medium\), hatching \(texture\), PVDC, hatching style):6.2, BREAK (masterpiece), (best quality), (detailed), (8k,Ultra HD), (high quality, 8k hd, absurd res, very awa, very aesthetic, newest), dynamic angle, BREAK 1girl, female, solo, smoking, cigarette, mouth_hold, blood on face, sketch, aged up, upper body, mafia, simple background, white background, sketch, illust, depth of field, masterpiece, best quality, amazing quality, very aesthetic, absurdres,
Negative prompt
(realistic, photorealistic, hyperrealistic, 3d, 2.5d, cgi, flat color:1.2), (bad quality, low quality, worst quality, bad anatomy, extra fingers, distorted hands:1.3), (deformed, mutated, blurry:1.1),
Settings
Sampler
Euler a
Steps
24
Guidance
4
Seed
1597979679
Size
832x1216
Clip skip
2
Prompt
wlop, (ssambatea, quasarcake, rella:0.825), (gogalking, krekkov:0.75), ebifurya, 1girl, upper body, white hair, albino, makeup, blush, eyeshadow, pink choker, colored eyelashes, hair between eyes, hair over eyes, sidelocks, low tied hair, hair over shoulder, long hair, messy hair, long sleeves, pale skin, red eyes, detailed eyes, white sundress, simple background, black background, sleeves past wrists, sleeves past fingers, white theme, masterpiece, best quality, amazing quality, very awa, absurdres, very aesthetic chiaroscuro:1.8
Negative prompt
bad quality,worst quality,worst detail,sketch, censor, patreon username, patreon logo, watermark, artist name, signature, logo, anatomical nonsense, bad anatomy, ugly, worst aesthetic, old, mismatched pupils, dirty face, extra limbs, bad hands, bad fingers, text, normal quality, low quality, web address, email address,
Settings
Sampler
DPM++ 2M
Steps
16
Guidance
5.497
Seed
3926965926
Size
832x1216
Prompt
(masterpiece, masterwork, best quality, amazing quality, very aesthetic, ultra-detailed, high detail, detailed, absurdres, superabsurd res, high resolution, hi res, HDR, 8K, 4K, Ultra HD, newest, beautiful, beautiful face, cute girl, very awa, absolutely eye-catching, extremely detailed eyes, EyesHD:1.2, clear skin, eyelashes, sharp focus, depth of field, high contrast, brightly lit, highly visible, perfect lighting, best lighting, volumetric lighting, light rays, ray tracing, raytracing, ambient occlusion, soft lighting, cinematic, cinematic composition, epic, detailed background, intricate cinematic background, detailed illustration, scenery, illustrating, CG, painting_(artwork), painting \(medium\), oil painting \(medium\), pastel \(medium\), hatching \(texture\), fabulosity style, chiaroscuro, pvdc_il:3.5, PVDC_STL:3.5, hatched style:3.5, sketch style:3.5, thick lines:3.5, (((pvdc_il, pvdc_stl, hatched style, thick lines)))), colorful ink splash, intense expression, black and white theme, tenebrism, blurry foreground, split theme, (close-up:1.5), poster \(medium\), artbook, BREAK kochiya sanae, 1girl, solo, intricate details, masterpiece, best quality, absurdres, detailed face, detailed eyes, skindentation, blurry background, , , anime coloring, performance uniform, stage with light beams, cleavage window, ribbon twintail, playful , arched back, soft smile, glistening skin, patterned lighting, upper body, cropped, BREAK (qiandaiyiyu:0.85),(soleil \(soleilmtfbwy03\):0.6),(godiva ghoul:0.65),(anniechromes:0.5), (rella:1.2),(redum4:1.2) (au \(d elete\):1.2) (dino \(dinoartforame\):1.1),wlop,
Negative prompt
low quality,bad quality,worst quality,worst detail, lowres, (low quality:1.2), censored, mosaic censoring, anatomical nonsense, bad anatomy, worst aesthetic, old, mismatched pupils, extra limbs, bad hands, deformed hands, missing finger, bad fingers, watermark, artist name, signature, logo, patreon logo, patreon username, web address, email address,
AI Models for Image, Video & Text Generation
velvette4 xl is a base model 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