CivArchive
    Kitten Tower - v1.0 (v-pred)
    NSFW
    Preview 54583317
    Preview 54583528
    Preview 54583750
    Preview 54584006
    Preview 54584306
    Preview 54584329
    Preview 54584433
    Preview 54584642
    Preview 54584683
    Preview 54584694

    Kitten Tower is NoobAI XL model focused on cute style.

    VAE is baked in.

    You need to use the webui that support v-prediction.

    • ComfyUI

    • reForge

    • AUTOMATIC1111 (place the config file in the same folder as the checkpoint)

    License

    This model adheres to the NoobAI-XL License (Modified Fair AI Public License 1.0-SD) and is available for non-commercial use only. This license prohibits any form of commercialization, including but not limited to monetization or commercial use of the model, derivative models, or model-generated products.

    • Steps: 25-30

    • CFG scale: 5-7

    • Sampler :Euler a, Euler

      I recommend using "Euler" if you use v-pred model. Using "Euler a" sometimes generates colored skin and objects such as red, blue and so on.

    • Scheduler for vPred : Normal, Simple (Other schedulers produce corrupt images)

    • Use ADetailer as needed

    Positive Prompt

    masterpiece, best quality

    Negative Prompt

    worst quality, bad quality, low quality, lowres, scan artifacts, jpeg artifacts, sketch, light particles

    Merge recipes (Epsilon-pred v1.0)

    1. Merging the difference between CyberRealistic XL v5 and stable-diffusion-xl-base-1.0 to NoobAI-XL Epsilon-pred v1.1 by "Add Difference" (Checkpoint A)

    2. Replacing the CLIP of Checkpoint A by NoobAI-XL Epsilon-pred v1.1 (Checkpoint B)

    3. Merging Cat Carrier v3.0 to Checkpoint B by "Weighted Sum" with block (0,0,0,0,0,0,0,0,0.3,0.3,0.3,0,0.3,0.3,0.3,0,0,0.3,0,0) (Checkpoint C)

    4. Merging Cat Tower Epsilon-pred v1.2 to Checkpoint C by "Add Cosine B" with block (0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0,0,0,0,0) (Checkpoint D)

    5. Merge the difference between ioliPonyMix v3.1 and ponyDiffusionV6XL to Checkpoint D by "Train Difference" with block (0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0,0,0,0,0) (Checkpoint E)

    6. Merge the difference between IllusionBreed v3.0 and NoobAI-XL-Epsilon-pred 1.1-Version to Checkpoint E by "Train Difference" with block (0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.0,0,0,0,0) (Checkpoint F)

    7. Merge the difference between CatPod v1.0(I merged it but not published) and ponyDiffusionV6XL to Checkpoint F by "Train Difference" with block (0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0,0,0,0,0) (Checkpoint G)

    8. Merge style LoRA to Checkpoint G (Kitten Tower Epsilon-pred v1.0)

    Merge recipes (vPred v2.0)

    1. Merging the difference between CyberRealistic XL v5 and stable-diffusion-xl-base-1.0 to NoobAI-V-Pred-1.0-Version by "Add Difference" (Checkpoint A)

    2. Replacing the CLIP of Checkpoint A by NoobAI-V-Pred-1.0-Version (Checkpoint B)

    3. Merging Cat Carrier v3.0 to Checkpoint B by "Weighted Sum" with block (0,0,0,0,0,0,0,0,0.4,0.4,0.4,0,0.4,0.4,0.4,0,0,0.4,0,0) (Checkpoint C)

    4. Merging Kitten Tower vpred v1.0 to Checkpoint C by "Add Cosine B" with block (0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0,0,0,0,0) (Checkpoint D)

    5. Merge the difference between ioliPonyMix v3.1 and ponyDiffusionV6XL to Checkpoint D by "Train Difference" with block (0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0,0,0,0,0) (Checkpoint E)

    6. Merge the difference between IllusionBreed v3.0 and NoobAI-XL-Epsilon-pred 1.1-Version to Checkpoint E by "Train Difference" with block (0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.0,0,0,0,0) (Checkpoint F)

    7. Merge the difference between CatPod v1.0(I merged it but not published) and ponyDiffusionV6XL to Checkpoint F by "Train Difference" with block (0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0,0,0,0,0) (Checkpoint G)

    8. Merge style LoRA to Checkpoint G (Checkpoint H)

    9. Add v_pred and ztsnr keys to Checkpoint H by noob_v_pencil-XL merge recipes (Kitten Tower vPred v1.0)

    Merge recipes (vPred v1.0)

    1. Merge Cat Tower v1.5 vPred to NoobAI-V-Pred-1.0-Version by "Weighted Sum" with block (0,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0,0,0,0,0) (Checkpoint A)

    2. Merge the difference between ioliPonyMix v3.1 and ponyDiffusionV6XL to Checkpoint A by "Train Difference" with block (0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0,0,0,0,0) (Checkpoint B)

    3. Merge the difference between IllusionBreed v3.0 and NoobAI-XL-Epsilon-pred 1.1-Version to Checkpoint B by "Train Difference" with block (0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.0,0,0,0,0) (Checkpoint C)

    4. Merge the difference between CatPod v1.0(I merged it but not published) and ponyDiffusionV6XL to Checkpoint C by "Train Difference" with block (0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0,0,0,0,0) (Checkpoint D)

    5. Merge style LoRA to Checkpoint D (Checkpoint E)

    6. Add v_pred and ztsnr keys to Checkpoint E by noob_v_pencil-XL merge recipes (Kitten Tower vPred v1.0)

    Description

    FAQ

    Comments (7)

    hboxgames132Jan 30, 2025· 3 reactions
    CivitAI

    looks very cute

    anhapiai567Feb 2, 2025
    CivitAI

    VAEは何を使いますか?あなたの画像のステータスだとノイズだらけになってしまします

    nuko_masshigura
    Author
    Feb 2, 2025

    @anhapiai567 VAEはモデル内蔵のものでいいです。このモデルはv-predictionを使用しているので、v-predictionをサポートしているクライアントを使用する必要があります。https://www.ipentec.com/document/ai-image/stable-diffusion-noise-image-generated-using-v-prediction-modelを参考にしてください。

    anhapiai567Feb 3, 2025

    @nuko_masshigura 質問すいません、ReForgeとForgeは同じですか?Forgeを使用してるのですが[Advanced Model Sampling for reForgeが無いです

    nuko_masshigura
    Author
    Feb 3, 2025· 1 reaction

    @anhapiai567 reForgeとForgeは別物です。Advanced Model Sampling for reForgeはreForgeにしかありません

    petrouchkak651Apr 19, 2025· 1 reaction
    CivitAI

    Any possibility for next update?

    It's indeed a great model for cute characters

    Just not good for shaped pupils(cross shaped, star shaped, flower shaped .etc), and may generate extra digits on hands and feet.

    nuko_masshigura
    Author
    Apr 20, 2025· 1 reaction

    @petrouchkak651 Yes, I'll update Cat Tower, the base model for this model, and then I'll get to work on this one.