My attempt at creating LoRA with the style of artist GarikAliev.I am a beginner so this is probably quite clumsy, hopefully in the future I will be able to improve it. To train this LoRA I used mostly his old works, taken from open sources (Rule 34 site mostly).All rights to the materials on which this model was trained belong to the original artist (Garikaliev), he is easy to find by nickname, support him if you want, without him this LoRA would definitely not exist, plus, for now it just imitates the original style rather than reproduces it.I recommend you to use Sampler Euler a since it generally makes the generation cost a little lower.I also recommend you add clarifications about the number of characters to the prompt, if there is only one, then add the words "Solo, alone", I also recommend adding clarifying hints like "five fingers on hands", this can prevent some defects.Best suited for creating cartoon bara works, with anime and cartoon characters.

GarikAliev style

Different base model (Illustrious instead of Pony), more images used to train the model.

GarikAliev_style_V2-000013.safetensors

V2 (Illustrious)

V3 (NoobAi)

V3 (Illustrious)

V1 (Pony)

<span style="background-color:rgb(26, 27, 30);color:rgb(193, 194, 197);font-family:-apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, Helvetica, Arial, sans-serif, &quot;Apple Color Emoji&quot;, &quot;Segoe UI Emoji&quot;;font-size:16px">LoRA with the style of artist GarikAliev. The first version was based on Pony, this second one is based on Illustrious.

GarikAliev style V2

Another base model (Illustrious instead of Pony).The number of pictures used for training is twice as large this time.

Files with SHA256: 61b8b68d30dd6241e52a63ae0124ab15a514473fd42f3fb78bb5b5712fe38647