CivArchive
    Grace Boor - v5.0
    Preview undefined
    Preview undefined

    Please feel free to make any request for Embeddings you would like me to make. I don't charge for them. Also, please leave review images. I love seeing how people use my models :D

    Grace Boor: a 22 year old American fashion model/Onlyfans creator

    I want to thank Denche354 for motivating me to finally learn how to make Embeddings. This my first embedding. Planning on making more/improving Embeddings in the future. I would love your feed back.

    Description

    FAQ

    Comments (4)

    WoBNov 11, 2024· 1 reaction
    CivitAI

    Very nice. Did you just experiment until you found good settings or did you follow a good guide? I still haven't been able to make an embedding that I would say is good yet.

    Twinmind00
    Author
    Nov 11, 2024· 1 reaction

    So I had been trying to make embeddings for months, but everything was coming out wrong, so I had kind of given up. I had been doing review images of embeddings from Denche354 (a great guy btw). We got to talking and he linked me to this Embedding Guide. After following that guide, everything worked well. I did refine the process a bit.

    7 things I would add/change from the guide.

    1. Closeup of faces are far more important than diverse setting or body shots.

    2. Less bare skin is better. SD likes to take clothing off by default.

    3. Try and find images of them looking directly at the camera an not profile shots.

    4. Try and find images depicting a wide variaty of emotions if possible.

    5. I use Microsoft photos to correct color, tint, saturation, etc. It even has an A.I. erase tool now.

    6. After generating the BLIP texts during batching, make sure to open them up and correct any mistakes.

    7. 3000 steps seems to be ideal instead of the 1500 that is recommended.

    WoBNov 12, 2024· 1 reaction

    @Twinmind00 Thanks very much! I'll try out that guide and your additions. I don't have many different expression images, but I do have a tool on Pinokio that you can change up the face's expression, so I might try that on top of it.

    How many images do you generally try to gather for each embed?

    Twinmind00
    Author
    Nov 12, 2024· 1 reaction

    @WoB 20 images. The expression thing isn't required. Just helps. Also if you need any additional help during the process feel free to message me directly using civitai's chat system.

    TextualInversion
    SD 1.5

    Details

    Downloads
    127
    Platform
    CivitAI
    Platform Status
    Deleted
    Created
    11/5/2024
    Updated
    5/7/2026
    Deleted
    5/23/2025
    Trigger Words:
    GraceBoorV5_TMs

    Files

    GraceBoorV5_TMs.pt

    Mirrors

    HuggingFace (1 mirrors)
    CivitAI (1 mirrors)

    Available On (1 platform)

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