language:
- en
tags:
- stable-diffusion
- text-to-image
license: creativeml-openrail-m
inference: true
pipeline_tag: text-to-image
i modelli 'Unico'
Unico is the series of custom mixed models. Based on Inizio Unico and AbyssOrange2 models with U-Net Merge, and support .safetensors format only.
WebUI-amicable
Summary
This model repository includes 4 main models currently:
Model: A Model: B Merge Weight Base alpha Merge Name Inizio Fantasma+Inizio Inseguitore+Inizio Foschia Inizio Replicante+Inizio Skinjob+Inizio Deckard weighted, M=0.66666666+M=0.66666666 N/A Unico
Unico is another form of Inizio Unico.
Model: A Model: B Merge Weight Base alpha Merge Name Inizio Unico AbyssOrange2 SFW weighted, M=0.75. N/A Unico Arancia
Unico Arancia('Orange🍊') is the closest model from AbyssOrange2 SFW.
Model: A Model: B U-Net Merge Weight Base alpha Merge Name Unico Arancia (Openniji 0,0,0,0,0,0,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 0 Unico Arancia 1.3
Unico Arancia('Orange🍊') 1.3 is improved model of Unico Arancia. Anime~Semi-realistic
Model: A Model: B U-Net Merge Weight Base alpha Merge Name Unico Arancia (Openniji 0,0,0,0,0,0,1,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0 0 Unico Arancia Speciale
Unico Arancia Speciale('Orange🍊 Special') is specially improved model of Unico Arancia to increase delicate details.
Model: A Model: B U-Net Merge Weight Base alpha Merge Name (Unico Arancia+Unico Pera+Unico Cartamo+Unico Castagna+Unico Vaniglia) (Elldreth 4060+SamDoesArtsUltMerge) experimental mix N/A Unico Castagna 2.0
Unico Castagna('Chestnut🌰') 2.0 is the improved model which focused to Semi-realistic only.
- NOTE: Another models are moved to legacy folder.
Setting Reccomendation
##Basic prompts for anime
"txt2img/Prompt/value": "(best quality, extreme intricate detailed, octane render, very delicate cinematic light, colourful), (/*place tags*/), (solo girl/*character tags*/), (/*pose tags*/), (big breasts, big pelvis, slim waist, long legs, best ratio four finger and one thumb, /*body tags*/), (beautiful eyes and face), (/*colour of hair tag*/ hair, /*colour of eyes*/ eyes, thick lips, lip gloss), (/*clothing tags*/)",
"txt2img/Negative prompt/value": "(nsfw, worst quality, low quality:1.4), (greyscale), (fingers(missing, fused, interlocked, abnormal, too many, bad anatomy, fused, fusion, lose, bad detailed, mutated), digit(extra, fewer), hands(greater than 4 fingers, less than 4 fingers, cropped, mutated):1.4), (fat, chubby, curvy, watermark, signature:1.4), (3d, realistic)"
- Variational Automatic Encoder: SD MSE 840k.vae.safetensors
- Clip Skip: 2
- Resolution: 1024x576 w/ HighRes. Fix
- HighRes. Fix: R-ESRGAN General WDN 4xV3; upscale by 1.25
Sample Images
▲ X/Y Plot #1
▲ Unico Arancia
License Information
This model follows Creative ML Open RAIL-M: Stable Diffusion License But, You may use whatever you want. I don't like to set such restriction.
Contact
[email protected] or Find Cinnamomo on Arcalive AI Art Channel