Model
Trends
.ai
Model
Trends
.ai
the best open source ai models
Run Locally
ng_deepnegative_v1_75t.pt
0 MB · PickleTensor
ng_deepnegative_v1_75t.safetensors
0 MB · SafeTensor
Download
Downloads are served by Civitai. Some creators ask for a Civitai login first.
AD
Install / Run on Graydient
Requires membership
<deepnegative-v1x-xl>
About This Model
STABLE DIFFUSION XL 1.0 · Embeddings
Created by
lenML
on Civitai
by lenML - Deep Negative V1.x
Creator notes
This embedding will tell you what is REALLY DISGUSTING🤢🤮 So please put it in negative prompt😜 ⚠This model is not trained for SDXL and may bring undesired results when used in SDXL. If you use SDXL, recommended this 👇 another deep-negative: - pony version: - SDXL version: TOP Q&A - how to use TI model? - what is negative prompt? [Special Reminder] If your webui reports the following errors: - CUDA: CUDA error: device-side assert triggered - Assertion -sizes[i] The reason is that many scripts do not handle overly long negative prompt words (greater than 75 tokens) properly, so choosing a smaller token version can improve this situation. [Update:230120] What does it do? These embedding learn what disgusting compositions and color patterns are, including faulty human anatomy, offensive color schemes, upside-down spatial structures, and more. Placing it in the negative can go a long way to avoiding these things. - What is 2T 4T 16T 32T? Number of vectors per token [Update:230120] What is 64T 75T？ 64T: Train over 30,000 steps on mixed datasets. 75T: embedding limit maximum size, training 10,000 steps on a special dataset (generated by many different sd models and special reverse processing) Which one should choose? - 75T: The most ”easy to use“ embedding, which is trained from its accurate dataset created in a special way with almost no side effects. And it contains enough information to cover various usage scenarios. But for some "good-trained-model" may hard to effect and, change about may be subtle and not drastic enough. - 64T: It works for all models, but has side effect. so, some tuning is required to find the best weight. recommend: [( NG_DeepNegative_V1_64T :0.9) :0.1] - 32T: Useful, but too more - 16T: Reduces the chance of drawing bad anatomy, but may draw ugly faces. Suitable for raising architecture level. - 4T: Reduces the chance of drawing bad anatomy, but has a little effect on light and shadow - 2T: ”easy to use“ like T75, but just a little effect Suggestion Because this embedding is learning how to create disgusting concepts, it cannot improve the picture quality accurately, so it is best used with (worst quality, low quality, logo, text, watermark, username) these negative prompts. Of course, it is completely fine to use with other similar negative embeddings. More examples and tests - draw building: - hand fix: - portrait (with ): - fusion body fix: How is it work? I tried to make SD learn what is really disgusting with deepdream algorithm, the dataset is imagenet-mini (1000 images chosen randomly from the dataset again) deepdream is REALLLLLLLLLLLLLLLLLLLLLY disgusting 🤮 and process of training this model really made me experience physical discomfort 😂 Backup
Trigger words
ng_deepnegative_v1_75t
493,041 downloads · 22,090 likes on Civitai
SHA256
54E7E4826D53949A3D0DDE40AEA023B1E456A618C608A7630E3999FD38F93245
ModelTrends.ai Model ID
#15511
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
19
Guidance
4
Seed
1737173171
Size
832x1216
Clip skip
2
Weight used
1
Prompt
score_9, score_8_up, score_8, (solo), (solo:2), (fluffy:0.8), (feral), outdoors, furry, (((mouse))), female, bright green eyes, grey fur, grey ears, claws, (fluffy cheeks:0.4), grey hooded cape, hood up, (blue dress), cute, in the style of beatrix potter, cel-shading, fca style, village, night, rain, (dark:1.5), backlit, (side view:1.3), (upper body shot), look back
Negative prompt
ng_deepnegative_v1_75t, score_6, score_5, score_4, (cub, young, child, chibi:1.5), busty, ugly face, low res, blurry face, ugly hands, pumped body, black and white, penis, clothing, breasts, big ears, ((hair)), nipples, green fur, belly, extra ears, human, blue fur, panties, simple background, signature
Settings
Sampler
DPM++ 2M Karras
Steps
20
Guidance
7
Seed
2676625826
Size
512x768
Prompt
(masterpiece:1.2), (photo of a partially buried cyborg, sandblasted and rusted, their robotic arm reaching towards a flickering sun Emphasizing the harsh environment, with cracked lenses and fading paint details:1.1), (bathed in the cold light of a holographic moon:1.1), (art by Cyril Rolando:1.4), (close-up shot:1.0), (hyperdetailed:1.1), (intricate details:1.0), (Refined details:1.1), (best quality:1.1), (high resolution:1.2)
Negative prompt
embedding:EasyNegative, embedding:negative_hand-neg, (worst quality, low quality, bad quality:1.3), low-res, blur, blurry, (extra fingers, bad anatomy, wrong proportions, deformed, mutated hands and fingers:1.2), (monochrome, grayscale:1.1), image artifacts, low resolution, (watermark:1.2), loli, headphones, text, embedding:ng_deepnegative_v1_75t, embedding:AloeCrapArt-5000
Settings
Sampler
Euler a
Steps
29
Guidance
4
Seed
4288902158
Size
832x1216
Clip skip
2
Weight used
1
Prompt
score_9, score_8_up, score_7_up, score_6_up, 1girl, solo, portrait, brown hair, lips, freckles, biting own lip, (biting lip:1), seductive, ponytail, lipstick, makeup, eyeliner, eyeshadow, beautiful detailed face, sexy woman, open shirt, no bra, night, night club background, blurry background
Negative prompt
score_6,score_5,score_4, worst quality, low quality, bad anatomy, bad hands, missing fingers, fewer digits, source_furry, source_pony, source_cartoon,3d, nipples, easynegative, ng_deepnegative_v1_75t, simple background, bad hands, missing fingers, extra fingers, deformed hands,
Settings
Sampler
Euler a
Steps
40
Guidance
4
Seed
338164970
Size
832x1216
Clip skip
2
Weight used
1
Prompt
top- down, fantastical and surreal scene, closeup inside of an enormous ancient pocket_clock standing upright with its cover open revealing a magical portal, portal inside the clock, Through the portal is a breathtaking landscape unfolds, towering mountains, a cascading volcano, a steaming river winding through a burned forests, The edges of the pocket_clock are adorned with intricate glowing runes and vines, the clock resting on obsidian floor, mystical touch, illuminated by a crescent moon and twinkling stars, dreamlike and otherworldly atmosphere, sense of adventure and wonder, ((red-blue, artifact, <lora:aidmaMTG-CardsXL-v0.1:1> ,mtg, mtg card)), view from above, (score_9, score_8_up:1.1), score_7_up, a fine color ink lineart sketch, (reflective volumetric), raytracing, absurdres, dark, low-key,
Negative prompt
boring, bad quality, (low quality, worst quality:1.4),(bad_prompt:0.8), (monochrome:1.1), negativeXL_D, aidxlv05_neg, jpeg artifacts, unsharp, ugly, deformed, noisy, score_6, score_5, score_4, cropped, lowres, strabismus, ugly, old, unrealistic physics, unrealistic, dry, bad background, distorted, not detailed, simple, cgi, distorted card,
Settings
Sampler
DPM++ 2M Karras
Steps
50
Guidance
3.5
Seed
2031206838
Size
832x1216
Clip skip
2
Weight used
1
Prompt
Craft a captivating vibrant dark black-neon caption for a photo capturing the essence of a cyborg Bedouin sorcerer in fight stance, defying expectations by conjuring a whimsical robot 'freak' from an oversized sedge hat. ethereal, smoky backdrop. throwing a translucent orange/tanslucent purple/black iaidow, weapon, katana, holding sword, ready to draw, sheathed, unsheathing, scabbard,, very detailed, atmospheric haze, Film grain, cinematic film still, shallow depth of field, highly detailed, high budget, cinemascope, moody, epic, OverallDetail, gorgeous, 2000s vintage RAW photo, photorealistic, candid camera, color graded cinematic, eye catchlights, atmospheric lighting, skin pores, imperfections, natural, shallow dof
Negative prompt
signature, text, easynegative, bad proportions, low resolution, bad, ugly, terrible, render, watermark, signature, worst quality, low quality, normal quality, lowres, simple background, inaccurate limb, extra fingers, fewer fingers, missing fingers, extra arms, (extra legs:1.3), inaccurate eyes, bad composition, bad anatomy, error, extra digit, fewer digits, cropped, low res, worst quality, low quality, normal quality, jpeg artifacts, extra digit, fewer digits, trademark, watermark, artist's name, username, signature, text, words, human,
Settings
Sampler
DPM++ 2M
Steps
32
Guidance
4
Seed
149671720
Size
832x1216
Clip skip
2
Weight used
1
Prompt
safe_pos, score_9, score_8_up, score_7_up, pov from the side, dynamic pose, beautiful, pink hair, female elf, glowing green eyes, long hair, small breasts, pink lipstick, dark black leather armor, holding enchanted shortsword, dungeon background, BREAK (((fully clothed)))
Negative prompt
safe_neg, simple background, bad anatomy. watermark, easynegative, score_6, score_5, score_4, bad quality, ugly face, ugly style, bad eyes, mutated hands, bad hands, worst quality , censored, high contrast , sketch , monochrome , furry, signature
AI Models for Image, Video & Text Generation
deepnegative v1x xl is a embedding 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