CivArchive
    noob_v_pencil-XL - v0.5.1
    NSFW
    Preview 1Preview 2Preview 3Preview 4Preview 5Preview 6Preview 7Preview 8Preview 9Preview 10

    Noobier than NoobAI-XL. (but difficult to get started with)
    Notice: This is a v-prediction model. Read "How to Use the Model" section on NoobAI-XL model page.


    License: NoobAI-XL License (Modified Fair AI Public License 1.0-SD)
    You MUST share the merge recipe if you release a model merged with noob_v_pencil-XL.
    This license prohibits any form of commercialization.
    For more details, please see the license section of NoobAI-XL model page.

    Please see the "About this Version" section for information on each version.


    Sampling method: Euler /Euler a (Other sampling methods are not supported by this v-prediction model.)
    Schedule Type: Automatic
    CFG: 4 - 5
    Prompt: masterpiece, best quality, newest
    Negative Prompt: worst quality, bad quality, old, early

    Description

    NoobAI-XL (NAI-XL) V-Pred-0.75S-Version Based

    # Tools

    # Checkpoints

    # Recipe

    1. Merge using sd-mecha for ComfyUI.

    # sd-mecha format recipe
    model "sdxl\noob_v_pencil-XL-v0.5.0.safetensors" "sdxl" "base"
    model "sdxl\noob_v_pencil-XL-0.75s.safetensors" "sdxl" "base"
    model "sdxl\noobaiXLNAIXL_epsilonPred10Version.safetensors" "sdxl" "base"
    merge "clamped_add_opposite" &0 &1 &2 alpha=1.0
    model "sdxl\noob_v_pencil-XL-v0.3.0.safetensors" "sdxl" "base"
    model "sdxl\noobaiXLNAIXL_vPred065SVersion.safetensors" "sdxl" "base"
    merge "clamped_add_opposite" &3 &4 &5 alpha=1.0
    model "sdxl\noobaiXLNAIXL_vPred075SVersion.safetensors" "sdxl" "base"
    merge "add_cosine_b" &6 &7 alpha=0.05
    1. Add v_pred and ztsnr keys.

    # Python Script
    BASE_MODEL_NAME = "noob_v_pencil-XL-v0.5.1"
    tensors = {}
    
    with safe_open(BASE_MODEL_NAME + "-base.safetensors", framework="pt", device="cpu") as f:
      for key in f.keys():
        tensors[key] = f.get_tensor(key)
    
    tensors["v_pred"] = torch.tensor([0.0])
    tensors["ztsnr"] = torch.tensor([0.0])
    save_file(tensors, BASE_MODEL_NAME + ".safetensors")