Model
Trends
.ai
Model
Trends
.ai
the best open source ai models
Run Locally
coma_v2.safetensors
2.14 GB · SafeTensor · fp16 · pruned
Downloads are served by Civitai. Some creators ask for a Civitai login first.
About This Model
STABLE DIFFUSION 1.5 · Base Models
Created by
UMAZTLI
on Civitai
Coma by skumerz. This model has a certain style to it, kinda like niji and kind of like anime? There are 3 versions: Sfw, Nsfw, Hard
Creator notes
Hello thank you for taking an interest in my model ♥ I do plan to update this : ) Version2 made it more simple and more 2d. can do NSFW baked in VAE Tools used: Easynegative - badhandv4 - Vae used - ((V2 already has baked in vae)) Upscaler used - -- 4x-UniScaleV2_moderate ((please use hire.fix to improve quality)) If necessary be sure to use inpaint after upscaling to polish things up even more, Be sure to use: (worst quality:2), (low quality:2), (normal quality:2), lowres, bad anatomy, normal quality, ((monochrome)), ((grayscale)) - in negative prompt to get similar results _______________________________________________________________________ 1st Versions I merged a bunch of models and also did some slight training with a lot of lora fine tuning. This model has a certain style to it, kinda like niji and kind of like anime? There are 3 versions Sfw Nsfw Hard SFW version - does not do nsfw very well, which is why I created the nsfw versions. Keep in mind the sfw version is the original and how the model was intended to be. I prefer the details, chaos and shininess of this version but the other version might be better for your work flow. NSFW version - a merge between sfw and hard, not as good at doing nsfw compared to the hard version, but it can do nsfw at times. It just does not listen as well when it comes to nsfw prompts. Hard version - the most nsfw version out of the three flavors. (it can also do sfw stuff pretty good, but not as chaotic as SFW version) - It's not perfect but can do some nsfw stuff ok -- better than sfw version of course. I have not tested out any loras out when using these models so I have no idea how they react. Would love to see some results Be sure to use - official art, unity 8k wallpaper, ultra detailed, beautiful and aesthetic, High quality, masterpiece, best quality - or something similar to get better results and for negative, use: (worst quality:2), (low quality:2), (normal quality:2), lowres, bad anatomy, normal quality, ((monochrome)), ((grayscale)) - or something similar ((I am sorry about hands and sometimes anatomy can get weird😆))
0 downloads · 904 likes on Civitai
SHA256
44A301AB6269C4FB2379CB9860DDB2DE6B222A837483F24DB4AA15AA60D4B95F
ModelTrends.ai Model ID
#919
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
DPM++ 2M Karras
Steps
30
Guidance
7
Seed
542122431781272
Size
512x768
Prompt
(masterpiece, best quality, detailed, aesthetic, ornate, pastel colors, shimmering), (esoteric), (solo), cowboy shot, priestess, closed eyes, (ethereal transparent), (tempestmagic:1.2), fantasy, wind swept hair, flowing, clouds, arcane power, engulfed in energy, surrounded by magic, <lora:TempestMagic-21:1.2> <lora:wine128dim-epoch-000008:0.7>
Negative prompt
(worst quality, low quality:1.2), text, font, logo, copyright, watermark, signature, (from side:1.2), wings, embedding:bad-hands-5
Settings
Sampler
DPM++ 2M Karras
Steps
22
Guidance
4
Seed
1681466048
Size
616x1352
Prompt
(ultra-detailed, masterpiece), best quality:1.6), (cinematic lighting, vibrant colors, sharp), (creek, shore, running water, ripples:1.2), (reflections on water), (smooth boulders), autumn trees, gentle breeze, sunny, white clouds, peaceful, (wide sky, large sky, blue sky:1.3), BREAK. (intricate details:1.4), view from above, closeup, focus on surface, (OverallDetail, CS-BBz:1.1), BREAK. <lora:Ghibli_v6:0.85> <lora:StrokeTexture:0.25>
Negative prompt
(Worst quality, bad quality, bad drawing:1.3), deformed, watermark, nsfw, bokeh, human, girl, character, mountains, blurry
Settings
Sampler
DPM++ 2M Karras
Steps
30
Guidance
7
Seed
853914980757096
Size
512x768
Prompt
(masterpiece, best quality, detailed, aesthetic, ornate, pastel colors, shimmering), (esoteric), (solo), cowboy shot, priestess, closed eyes, (ethereal transparent), (tempestmagic:1.2), fantasy, wind swept hair, flowing, clouds, arcane power, engulfed in energy, surrounded by magic, <lora:TempestMagic-21:1.2> <lora:wine128dim-epoch-000008:0.7>
Negative prompt
(worst quality, low quality:1.2), text, font, logo, copyright, watermark, signature, (from side:1.2), wings, embedding:bad-hands-5
Settings
Sampler
DPM++ 2M Karras
Steps
30
Guidance
7
Seed
819069632834293
Size
512x768
Prompt
(masterpiece, best quality, detailed, intricate, pastel colors, iridescent), (solo), dynamic angle, (1girl), warrior, white dress, corset, ornaments, gemstones, golden accents, pauldrons, gauntlets, armor, blue hair, pointy ears, outdoors, castle, cloak, windy, night, clouds, wrenchsfantasy, fantasy, glowing <lora:DarkAndLight:0.7> <lora:wrenchsfantasy-000016:0.8>
Negative prompt
(worst quality, low quality:1.2), text, font, logo, copyright, watermark, signature, (side view), wires, cables, telephone pole, (hand), embedding:bad-hands-5
Settings
Sampler
DPM++ 2M Karras
Steps
28
Guidance
7.5
Seed
2598343992
Size
528x760
Clip skip
2
Prompt
menera style, by Ernst Haeckel, exquisite illustration of a girl wearing flowers, anime, pixiv, kawaii girl, facing forward
Negative prompt
(worst quality:2), (low quality:2), (normal quality:2), lowres, bad anatomy, normal quality, ((monochrome)), ((grayscale)), ((text, font, logo, copyright, watermark:1)), easynegative, badhandv4
Settings
Sampler
DPM++ 2M Karras
Steps
26
Guidance
6.5
Seed
3229412998
Size
528x760
Clip skip
2
Prompt
knight, a red aura violently erupts from the whole body, with black big sword targeting the planet in the sky, background open field, full body, masterpiece, V0id3nergy, stars in the background, demon horns
Negative prompt
(worst quality:2), (low quality:2), (normal quality:2), lowres, bad anatomy, normal quality, ((monochrome)), ((grayscale)), ((text, font, logo, copyright, watermark:1)), easynegative, badhandv4
Settings
Sampler
DPM++ 2M Karras
Steps
30
Guidance
7
Seed
897745477351511
Size
512x768
Prompt
(masterpiece, best quality, detailed, intricate, pastel colors, iridescent), (solo), dynamic angle, (1girl), warrior, white dress, corset, ornaments, gemstones, golden accents, pauldrons, gauntlets, armor, blue hair, pointy ears, outdoors, castle, cloak, windy, night, clouds, wrenchsfantasy, fantasy, glowing <lora:DarkAndLight:0.7> <lora:wrenchsfantasy-000016:0.8>
Negative prompt
(worst quality, low quality:1.2), text, font, logo, copyright, watermark, signature, (side view), wires, cables, telephone pole, (hand), embedding:bad-hands-5
Settings
Sampler
DPM++ 2M Karras
Steps
30
Guidance
7
Seed
910730575835034
Size
512x768
Prompt
(masterpiece, best quality, detailed, aesthetic, ornate, pastel colors, shimmering), (esoteric), (solo), cowboy shot, priestess, closed eyes, (ethereal transparent), (tempestmagic:1.2), fantasy, wind swept hair, flowing, clouds, arcane power, engulfed in energy, surrounded by magic, <lora:TempestMagic-21:1.2> <lora:wine128dim-epoch-000008:0.7>
Negative prompt
(worst quality, low quality:1.2), text, font, logo, copyright, watermark, signature, (from side:1.2), wings, embedding:bad-hands-5
AI Models for Image, Video & Text Generation
coma2 is a base model 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