metadata
language:
- en
tags:
- stable-diffusion
- text-to-image
license: creativeml-openrail-m
inference: true
pipeline_tag: text-to-image
i modelli 'Inizio'
Inizio is the series of custom mixed models. Based on many released open-source models, and support .safetensors format only.
WebUI-amicable
Summary
This model repository includes 7 models currently:
- Inizio Fantasma: Blossom mix+Anything 3.0+SamDoesArtsUltMerge; weighted, M=0.2. Quietly impressive semi-realistic model.
- Inizio Inseguitore: ALG+SamDoesArtsUltMerge+Blossom Mix; add difference, M=0.3. Similar to Inizio Fantasma, but anime-style focused.
- Inizio Foschia: ALG+Inizio Fantasma+A Certain Model; weighted, M=0.7. Similar to Inizio Inseguitore.
- Inizio Replicante: Inizio Foschia+DBmai+Finale 5o; weighted, M=0.5. Well-tuned Semi-Realistic Anime model. The most fashionable.
- Inizio Skinjob: Inizio Replicante+Berry Mix+ElyOrange; weighted, M=0.6. Well-tuned Semi-Realistic Anime model.
- Inizio Deckard: (Kribo Nstal+Inizio Skinjob; weighted, M=0.5)+Inizio Fantasma+Kribo Nstal; weighted, M=0.5.
- Inizio Unico Inizio Fantasma+Inizio Inseguitore+Inizio Foschia+Inizio Replicante+Inizio Skinjob+Inizio Deckard; weighted, M=1/6. The most advanced Inizio model.
Recommend Settings (Especially for Inizio Unico)
- Variable Automatic Encoder: SD MSE 840k.vae.pt
- Embedding:bad_prompt_ver2
- Clip skip: 3
- Prompt: object-related tag first, quality-related tag later; prompt list
- Resolution: 1024x576->1366x768 w/ HighRes. Fix
- HighRes. Fix: Latent or ESRGAN; upscale by 2
Sample Image
▲ X / Y Plot #1
▲ X / Y Plot #2
▲ X / Y Plot #3
▲ Inizio Skinjob
▲ Inizio Unico
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.