CivArchive
    Simple Anime - NSFW, Interpolation, Upscale, Lighting - v1.0
    NSFW
    Preview 103755248

    Hi! This is my first post on Civitai. I wanted to share my workflow for creating NSFW animations. I tried to keep it simple and organised so that the whole process is easy to understand.
    The zip file contains two versions: a normal one and a 2F2 (frame-to-frame) one. The latter allows you to create loops or longer animations by controlling the last frame.

    Description

    FAQ

    Comments (10)

    gxmattioli884Oct 2, 2025· 4 reactions
    CivitAI

    you gave mococo lore accurate mosquito bites XD

    TOF_enjoyerOct 6, 2025· 1 reaction
    CivitAI

    love this workflow, very simple and efficient. doesn't use 100 get and set node just to confuse you. 10/10

    guoywOct 7, 2025· 1 reaction
    CivitAI

    Thank you for helping me a lot

    brnyzychkDec 7, 2025
    CivitAI

    hi, wanna ask..

    is it using 2 models .safetensor and *.gguf ?

    NegativeMeta
    Author
    Dec 7, 2025

    yes it has the 2 options but it only use 1 at a time, you can use safetensor or gguf that are more lightweight, you only need to connect what format you want to use to the lora loader

    brnyzychkDec 8, 2025

    @NegativeMeta thx it works (im using first o last frame), but.. idk why the result was not very good compared to that im usually using
    in this workflow ksampler already determined to high 1 : low 2
    im usually using 1 : 1 on high low steps.. any advice?

    all lora and gguf used are same, but i notice the clip encoder was different

    brnyzychkDec 8, 2025

    @NegativeMeta idk what's wrong but seem i generate frame2frame, it always looping into the first frame (i already put different image on last frame tho)

    NegativeMeta
    Author
    Dec 9, 2025

    @brnyzych, hi you have to make sure that the second input image is active and connect its output to the sampler input (https://imgur.com/a/R0nLHIZ). The workflow has two functions: it can create loops using one image, with that image acting as the first and last, or it can create f2f video using a second image as the end.

    NegativeMeta
    Author
    Dec 9, 2025

    @brnyzychk did you still have this problem?, the moe ksampler is used to select the steps for high and low automatically, usually it uses 1:3 low:high

    brnyzychkDec 9, 2025

    @NegativeMeta i think it's fine now, already had fix on that.. anyway.. thanks :D

    Workflows
    Wan Video 2.2 I2V-A14B

    Details

    Downloads
    930
    Platform
    CivitAI
    Platform Status
    Available
    Created
    10/2/2025
    Updated
    5/12/2026
    Deleted
    -

    Files

    simpleAnimeNSFW_v10.zip

    Mirrors

    CivitAI (1 mirrors)