CivArchive
    Kelly Ripa Ti - v1.0
    NSFW
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined

    Kelly Maria Ripa is an American actress and television personality. Born in 1970, Ms. Ripa is known for her roles in All My Children, Hope & Faith, and as host of Live! with Kelly and Regis/Michael/Ryan/Mark.

    This embedding was trained on 178 distinct publicly-available images (128 of those repeated 3 times per epoch) yielding a total of 444 images processed per epoch with a batch size of 12 and 37 accumulation steps, running for a total of 40 epochs. All images in the training dataset were masked for learning attention; an ankle tattoo was removed on all source images. The training software OneTrainer was utilized; scheduler was ADAMW with a constant LR of 0.0006 (370 LR warmup steps).

    All sample images were generated using Highres fix. and ADetailer (face); all metadata prompts are accurate, though my seeds may be slightly different as all models used in the sample images are TRT/ONNX conversions (increases render speed). No additional LoRa, embedding or manual inpainting was utilized.

    I'd love to see generations leveraging this Ti, but please be respectful. Happy prompting!

    Description

    FAQ

    TextualInversion
    SD 1.5

    Details

    Downloads
    171
    Platform
    CivitAI
    Platform Status
    Deleted
    Created
    3/30/2024
    Updated
    7/7/2025
    Deleted
    5/23/2025
    Trigger Words:
    kellyRipa01

    Files

    kellyRipa01.safetensors

    Mirrors

    CivitAI (1 mirrors)
    TensorFiles (1 mirrors)