CivArchive
    [Style] Artist: Shikuro | iorimiyakiyo | いおり | @ioriran04 - v1.0
    Preview 143812341
    Preview 143812337
    Preview 143812343
    Preview 143812339

    Introduction

    This LoRA is trained on the illustration style of the artist ioriran04 / iorimiyakiyo.

    Usage

    Recommended LoRA strength: 1.0
    Trigger word: @shikuro (iorimiyakiyo)

    In ComfyUI, escape the parentheses:

    @shikuro \(iorimiyakiyo\)

    Training Details

    • Base model: Anima Base 1.0

    • Trainer: sorryhyun/anima_lora v1.17.1

    • Dataset: 62 captioned images

    • LoRA rank / alpha: 32 / 128

    • Learning rate: 2e-5

    • Optimizer: AdamW

    • Scheduler: Cosine, with 5% warmup

    • Batch size: 1

    • Training duration: 40 epochs / 2480 steps

    • Precision: BF16

    • Resolution: Aspect-ratio buckets at approximately 1 megapixel

    • GPU: NVIDIA RTX 3090

    简介

    这是基于 Anima Base 1.0 训练的 ioriran04 / iorimiyakiyo 画风 LoRA。

    使用

    推荐 LoRA 权重: 1.0

    激活词: @shikuro (iorimiyakiyo)

    在 ComfyUI 中需要转义括号:

    @shikuro \(iorimiyakiyo\)

    训练参数

    • 基础模型: Anima Base 1.0

    • 训练工具: sorryhyun/anima_lora v1.17.1

    • 数据集: 62 张带 caption 的图片

    • LoRA rank / alpha: 32 / 128

    • 学习率: 2e-5

    • 优化器: AdamW

    • 学习率调度: Cosine,5% warmup

    • Batch size: 1

    • 训练量: 40 epochs / 2480 steps

    • 训练精度: BF16

    • 训练分辨率: 约 1 M 像素的宽高比分桶

    • 显卡: NVIDIA RTX 3090

    Description

    FAQ

    LORA
    Anima

    Details

    Downloads
    176
    Platform
    CivitAI
    Platform Status
    Available
    Created
    9/25/2026
    Updated
    10/13/2026
    Deleted
    -
    Trigger Words:
    @shikuro \(iorimiyakiyo\)

    Files

    ioriran04-final.safetensors

    Mirrors

    CivitAI (1 mirrors)