CivArchive
    Catherine Zeta-Jones - The Mask of Zorro 1998 - v1.0
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined

    The Update of this model it's only available on Patreon:

    https://www.patreon.com/posts/97991250

    https://twitter.com/smoonfucker/status/1755000132427448498

    Trained with a lot of screenshots of the move [ The Mask of Zorro 1998 ].

    Check the metadata of the images posted as example.

    Use the next resolutions to get better results:

    • 512x712 [ For Portraits ]

    • 712x1024 [ For Full Body ]

    As a tip to correct face details and enhancements use ADetailer [ https://github.com/Bing-su/adetailer.git ] enhancer.


    https://www.patreon.com/smoonfucker

    Daily updated list of models: https://www.patreon.com/posts/list-of-models-89239085

    Please, you can also support me at:

    Thanks :_)

    Use of this model it's up to the responsibility of each user on posting any content, please avoid to post any NSFW content of the model to avoid model removal, users compliance or CIVITAI restrictions. Thanks!

    Description

    FAQ

    Comments (7)

    ReaperOfSoulsJun 6, 2023· 2 reactions
    CivitAI

    That's outstanding. Can you tell me how many images that you used in your dataset?

    smoonHacker
    Author
    Jun 7, 2023

    101 images, if I remember... I have a mess with folders with variations of resolutions, poses and pre-generated images used for training.

    Bit_ShifterJun 6, 2023· 1 reaction
    CivitAI

    Wow! Quite possibly a Lora with the best likeness I have seen! I have been attempting something similar using movie screenshots w/ mixed results. Would you mind sharing your training images and basic training settings (repeats/epochs)? I understand that it varies, but would give me ideas to try in the hope of achieving better results. I would be greatly appreciated!

    Great work!

    smoonHacker
    Author
    Jun 7, 2023

    The key here it's train your realistic models as LyCORIS/LoCon with more than 50 images, play around with the same config you use for your Loras.

    Bit_ShifterJun 9, 2023

    @smoonfucker Thank you. I understand most all the settings but was primarily interested in viewing the training images & captions you used for comparison with my own. I'm not getting anywhere near the level of consistency with likeness that this LoRA is generating and believe that the comparison would help me find the final piece of the puzzle I am missing.

    astragartistJun 7, 2023· 1 reaction
    CivitAI

    Congrats! I really appreciate a very hard work behind your model! I must test it in near future but I must comment that I like it very much!

    den1spla7er684Oct 5, 2023
    CivitAI

    Awesome work! Can you do the same with Antonio Banderas from the same movie? It would be great

    LoCon
    SD 1.5

    Details

    Downloads
    1,910
    Platform
    CivitAI
    Platform Status
    Deleted
    Created
    6/6/2023
    Updated
    7/7/2025
    Deleted
    5/23/2025
    Trigger Words:
    catherinezj

    Available On (1 platform)

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