CivArchive
    GhostbubblemangoV2 - GhostBubbleMango V2
    NSFW
    Preview 149646978
    Preview 149646975
    Preview 149647324
    Preview 149646957
    Preview 149646956
    Preview 149646961
    Preview 149648553
    Preview 149646959
    Preview 149646967
    Preview 149646970
    Preview 149646971
    Preview 149646973
    Preview 149646976
    Preview 149648554
    Preview 149646982
    Preview 149648126
    Preview 149648127
    Preview 149648130
    Preview 149648132
    Preview 149649032

    🥭 GhostBubbleMango V2 — Anima

    Version 2 of GhostBubbleMango!

    This version builds upon the original GhostBubbleMango merge with an updated model blend and an adjustment to the baked-in Turbo LoRA strength.

    For V2, the baked-in Turbo LoRA strength has been reduced from 1.0 to 0.5, with the goal of providing more flexibility in prompt interpretation, detail, and overall image generation while retaining the benefits of Turbo.

    This version combines GhostBubbleHybrid V3 and Pink Alchemy Anima, bringing together their respective styles into an updated GhostBubbleMango checkpoint.

    settings:
    CFG:1-3 Steps:12-20 Sampler: Euler A, er_sde

    What's changed in V2?

    • Updated model merge.

    • Baked-in Turbo LoRA strength reduced from 1.0 to 0.5.

    • Experimental adjustment intended to improve prompt flexibility and reduce some of the restrictions associated with stronger Turbo influence.

    • Updated blend of character styles and rendering characteristics.

    Source Models

    Notes

    This is an experimental update. Results may vary depending on prompt, sampler, steps, CFG, and LoRA combinations.

    Thanks for testing GhostBubbleMango, and I hope you enjoy V2!

    Description

    🥭 GhostBubbleMango V2 — Anima

    Version 2 of GhostBubbleMango!

    This version builds upon the original GhostBubbleMango merge with an updated model blend and an adjustment to the baked-in Turbo LoRA strength.

    For V2, the baked-in Turbo LoRA strength has been reduced from 1.0 to 0.5, with the goal of providing more flexibility in prompt interpretation, detail, and overall image generation while retaining the benefits of Turbo.

    This version combines GhostBubbleHybrid V3 and Pink Alchemy Anima, bringing together their respective styles into an updated GhostBubbleMango checkpoint.

    What's changed in V2?

    • Updated model merge.

    • Baked-in Turbo LoRA strength reduced from 1.0 to 0.5.

    • Experimental adjustment intended to improve prompt flexibility and reduce some of the restrictions associated with stronger Turbo influence.

    • Updated blend of character styles and rendering characteristics.

    Source Models

    Notes

    This is an experimental update. Results may vary depending on prompt, sampler, steps, CFG, and LoRA combinations.

    Thanks for testing GhostBubbleMango, and I hope you enjoy V2!

    Checkpoint
    Anima

    Details

    Downloads
    70
    Platform
    CivitAI
    Platform Status
    Available
    Created
    11/20/2026
    Updated
    11/20/2026
    Deleted
    -

    Files

    GhostbubblemangoV2.safetensors

    Mirrors