ComfyUI Node: CheckpointLoaderAdvancedDisTorch2MultiGPU

Authored by pollockjj

Created

Updated

475 stars

Category

multigpu/distorch_2

Inputs

ckpt_name
    unet_compute_device
    • cpu
    • cuda:0
    unet_virtual_vram_gb FLOAT
    unet_donor_device STRING
    clip_compute_device
    • cpu
    • cuda:0
    clip_virtual_vram_gb FLOAT
    clip_donor_device STRING
    vae_device
    • cpu
    • cuda:0
    unet_expert_mode_allocations STRING
    clip_expert_mode_allocations STRING
    high_precision_loras BOOLEAN

    Outputs

    MODEL

    CLIP

    VAE

    Extension: ComfyUI-MultiGPU

    This extension adds CUDA device selection to supported loader nodes in ComfyUI. By monkey-patching ComfyUI’s memory management, each model component (like UNet, Clip, or VAE) can be loaded on a specific GPU. Examples included are multi-GPU workflows for SDXL, FLUX, LTXVideo, and Hunyuan Video for both standard and GGUF loader nodes.

    Authored by pollockjj

    View Nodes

    Run ComfyUI workflows in the Cloud!

    No downloads or installs are required. Pay only for active GPU usage, not idle time. No complex setups and dependency issues

    Learn more