Nodes/ComfyUI-Omnivoice/OmniVoice Model Loader
ComfyUI Node

OmniVoice Model Loader

A ComfyUI node in OmniVoice with 3 inputs and 1 output.

By ethanfel·Created 5 months ago·Updated 5 months ago· 5
OmniVoice Model Loader
    • model
    devicecuda:0
    dtypefloat16
    compilefalse
    CategoryOmniVoice

    Inputs (3)

    NameTypeDefaultDescription
    deviceCOMBOcuda:03 options: cuda:0, cuda:1, cpu
    dtypeCOMBOfloat163 options: float16, bfloat16, float32
    compileBOOLEANfalseRun torch.compile() on the model after loading. First generation will be slow (~30-60s warmup) while the graph is compiled, then every subsequent generation in the session will be faster. Recommended for audiobook pipelines. Requires PyTorch 2.0+.

    Outputs (1)

    NameTypeDescription
    modelOMNIVOICE_MODEL