CivArchive
    2wink ensemble stars (Beta) - 2winkv1.1
    Preview 551693
    Preview 551692
    Preview 551691
    Preview 551689
    Preview 551695
    Preview 551814
    Preview 551813

    BETA MODEL

    This model is kinda hard to make because its hard to find good quality image of the twins.

    i will improve it later :DDDD

    this is my worst model so far 💀💀💀💀

    please use : https://huggingface.co/datasets/gsdf/EasyNegative
    and use higher fix on.


    Weights for lora: 0.6-1 please try what works best for you :DDD

    any comment, suggestions and tips are appreciated :D

    Description

    FAQ

    Comments (7)

    ditaApr 21, 2023
    CivitAI

    Hinata and Yuki both look really, really good! I would believe these were official artworks. You have even captured their personalities. I can tell you worked a lot on this one. Great job!

    shiowoneko
    Author
    Apr 21, 2023· 1 reaction

    thankyou :DD

    shiowoneko
    Author
    Apr 21, 2023
    CivitAI

    might take a 2 week break creating lora due to colabs new policy on web ui I am currently trying to learn the diffusers library or i could get a runpod

    ditaApr 21, 2023

    Colab decided to ban webui? I hadn't heard that. I guess because it makes it so easy to circumvent the "safety" check... 😩 Well... Maybe there will be another easy-to-use tool come along that is just as powerful as webui. (And then Google will ban it too…)

    shiowoneko
    Author
    Apr 21, 2023

    @dita don worry there are still diffusers but it will be a pain to learn. google is banning webui in the new tos for free users so i suggest learning diffusers 💀but making lora does not require you to have pro. anyways you can use TheLastBen colab on github that is still working for now

    ditaApr 26, 2023

    @shiowoneko Thanks for the info, shio. I've not looked in months now, but I don't recall ever seeing any good tools for using Diffusers efficiently. Of course, that doesn't really matter for Colab, since you run everything from a notebook anyway. But it still seems like a pain.

    I guess I should've started making LoRAs when I could do it more easily. There are still so many that I would like to have that I will have to make myself. Maybe one day...

    shiowoneko
    Author
    Apr 26, 2023

    @dita hey don't let that slow you down :DDDD from my experience using notebooks is easier since you don't need to mess with anything and just insert the image let it train. the hard part is finding the image that is good for training, here is the notebook i used if you want to try making your own :DDD https://github.com/Linaqruf/kohya-trainer

    or you can use this 1 click notebook:
    https://github.com/hollowstrawberry/kohya-colab

    LORA
    SD 1.5

    Details

    Downloads
    209
    Platform
    CivitAI
    Platform Status
    Available
    Created
    4/21/2023
    Updated
    6/28/2026
    Deleted
    -
    Trigger Words:
    2boys
    2wink
    smile
    twins
    brothers
    jacket
    orange_hair
    male_focus

    Available On (1 platform)

    Same model published on other platforms. May have additional downloads or version variants.