CivArchive
    Preview 144295855
    Preview 144295856
    Preview 144295853
    Preview 144295851
    Preview 144295850
    Preview 144295845
    Preview 144295844
    Preview 144295903
    Preview 144295891
    Preview 144295890
    Preview 144295887
    Preview 144295885
    Preview 144295882
    Preview 144295876
    Preview 144295877
    Preview 144295870
    Preview 144295868
    Preview 144295864
    Preview 144295861
    Preview 144295858
    • UPDATE 23/10/2026: Added new testing version for MOSS-Image, tested mostly with turbo lora. ER_SDE for more sharp images, Euler for more blurred.

    • UPDATE 30/09/2026:
      Added 2 more versions -
      Auto and Pro. I am testing new training methods, and while Adam is rather high quality, it breaks often and doesn't follow the style I trained it to follow that well. Auto is full on style, while Pro is blend between Auto and Adam, and breaks the least.

    Trigger words makes lora much stronger.

    Recommended settings for Krea2Turbo:

    Sampler:

    • er_sde | euler

    Scheduler:

    • beta | linear_quadratic | simple

    Lora Strength: 0.8-1.4 | Steps: 8-10 | CFG: 1

    Recommended settings for Anima:
    I recommend using v1 for softer look, v2 is good for upscaling or yottea's chromatic abberation.

    Sampler:

    • er_sde | euler | dpmpp_sde_gpu

    Scheduler:

    • beta | linear_quadratic | simple

    Lora Strength: 0.6-0.8 | Steps: 25 | CFG: 4

    Description

    FAQ

    LORA
    Krea 2

    Details

    Downloads
    102
    Platform
    CivitAI
    Platform Status
    Available
    Created
    9/30/2026
    Updated
    11/19/2026
    Deleted
    -
    Trigger Words:
    digital artwork
    anime-style art
    @96yottea

    Files

    YOTTEA-KREA-AUTO (1).safetensors

    Mirrors