CivArchive
    Rabbitadvisory Style for Noob/Illustrious - V-Pred-0.75S v0.1
    NSFW
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined

    Lora trained in the Rabbitadvisory's art style.

    Use about 350 images for training, and manually proofread the prompt words. Only contains furry, but can also produce non-furry images.

    Trigger Word:rabbitadvisory

    Weight: 0.6 - 1

    It is recommended to use this lora together with the improved quality lora.

    Support the Artist:https://rabbitadvisory.fanbox.cc/

    Enjoy.

    使用Rabbitadvisory风格训练的lora。

    使用大约350张手动校对标签的图像进行训练,只包含furry内容,但也能生成非furry。

    提示词:rabbitadvisory

    Weight: 0.6 - 1

    建议将该lora和提高质量的lora一起使用。

    请支持画师:https://rabbitadvisory.fanbox.cc/

    Description

    Noob V-pred 0.75S-Version

    FAQ

    Comments (10)

    AltacccountDec 13, 2024
    CivitAI

    What model was used for the reference pics?

    Flange
    Author
    Dec 16, 2024

    They are all generated using noobai models in comfyui, and the model version is the same named version of lora. Because the newer version of noobai is not in the association list of uploaded images, it cannot be associated.

    AFocJan 26, 2025
    CivitAI

    Sorry to bother but I think this lora is not working well on traditional ill models. Is there an ill version without v or epsilon-pred technique? Or could you post your dataset somewhere so that I can train on base model?

    Flange
    Author
    Jan 26, 2025· 1 reaction

    OK, I'll try to train a version on the base Illustrious model.

    AFocJan 26, 2025

    @Flange There are lots of splendid furry model these days like nova furry. Maybe they will have better performance and anatomical behaviors?

    AFocJan 26, 2025

    @Flange Oh I found you also trained yuio lora! It's one of my favorite artists too! We must share a common taste of art style!

    Flange
    Author
    Jan 27, 2025

    @laoxs That's great! In addition, when I tested the Illustrious model, I found that Noob seemed to have broader knowledge. For example, Noob could generate the correct lower body and soles of Cinderace without lora. I will slowly add the Illustrious V0.1 version of each lora later.

    AFocJan 28, 2025· 1 reaction

    @Flange Perhaps you could try the Nova and Mol_Keun models. They integrate some new knowledge and have a good anatomical performance. In fact, as long as it is not trained on models that are based on epsilon and v-prediction, the universality of the LoRA is generally better.

    AFocJan 28, 2025

    @Flange You might be interested in the artists like Kame3_t, Pepper0, Sparrowl 19, and Dont_jinxit. Some of them currently do not have ill versions available. If you are interested, you can give them a try!

    Flange
    Author
    Jan 29, 2025

    @laoxs I do like Kame's style and Sparrowl's style as well as his character Ivy. Very good suggestion, I will try to make it when I have time later.

    LORA
    NoobAI

    Details

    Downloads
    309
    Platform
    CivitAI
    Platform Status
    Available
    Created
    12/13/2024
    Updated
    6/12/2026
    Deleted
    -
    Trigger Words:
    rabbitadvisory

    Files

    Rabbitadvisory_Noob_V_Test1.safetensors

    Available On (2 platforms)

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