CivArchive
    Flash Photo SDXL - v1 - 800 steps
    NSFW
    Preview 2127192
    Preview 2127190
    Preview 2127466
    Preview 2127473
    Preview 2127482

    1 MB iA3 Network

    *UPDATE*: v6 is trained with Debiased Estimation Loss on the dev branch of kohya_ss sd-scripts. This makes image structure more coherent and have better color. If you train a model with Debiased Estimation Loss, you shouldn't need to use Min-SNR gamma or Color Aug as it does both better.

    Prompt format: very dark focused flash photo of <subject> at night, wide angle, black background

    I recommend using DPM++ 3M Exponential/Karras with 40-100 steps. Use the refiner at 0.8 for photorealism.

    Description

    1 MB iA3 Network

    Very WIP model focused on low-key flash photo images where the background is veiled in darkness.

    Prompt format:

    very dark focused flash photo of <subject> at night, wide angle, black background

    FAQ

    LORA
    SDXL 1.0

    Details

    Downloads
    549
    Platform
    CivitAI
    Platform Status
    Available
    Created
    8/20/2023
    Updated
    4/23/2026
    Deleted
    -
    Trigger Words:
    very dark focused flash photo

    Files

    sdxl_darkflash_V1-step00000800.safetensors

    Mirrors