ComfyUI Node: WanVideo Model Loader
Run WanVideo Model Loader without installing ComfyUI
This node already runs on ComfyICU. Open it in the browser, hit run, and you're billed for GPU seconds — not for setup, and not while you're idle.
Category
WanVideoWrapper
Inputs
model
base_precision
- fp32
- bf16
- fp16
- fp16_fast
quantization
- disabled
- fp8_e4m3fn
- fp8_e4m3fn_fast
- fp8_e4m3fn_scaled
- fp8_e4m3fn_scaled_fast
- fp8_e5m2
- fp8_e5m2_fast
- fp8_e5m2_scaled
- fp8_e5m2_scaled_fast
load_device
- main_device
- offload_device
attention_mode
- sdpa
- flash_attn_2
- flash_attn_3
- sageattn
- sageattn_3
- radial_sage_attention
- sageattn_compiled
- sageattn_ultravico
- comfy
compile_args WANCOMPILEARGS
block_swap_args BLOCKSWAPARGS
lora WANVIDLORA
vram_management_args VRAM_MANAGEMENTARGS
extra_model VACEPATH
fantasytalking_model FANTASYTALKINGMODEL
multitalk_model MULTITALKMODEL
fantasyportrait_model FANTASYPORTRAITMODEL
rms_norm_function
- default
- pytorch
Outputs
WANVIDEOMODEL
Extension: ComfyUI-WanVideoWrapper
ComfyUI wrapper nodes for a/WanVideo and related models.
Authored by kijai
Looking for a different node?
More nodes in ComfyUI-WanVideoWrapper
- Create CFG Schedule Float List
- Create Schedule Float List
- (Download)Load NLF Model
- (Down)load Wav2Vec Model
- Draw ArcFace Landmarks
- Draw Gaussian Noise On Image
- Draw NLF Poses
- Dummy Comfy Wan Model Object
- Extract Start Frames For Continuations
- Face Mask From Pose Keypoints
- FantasyPortrait Face Detector
- FantasyPortrait Model Loader
Other WanVideoWrapper nodes
Run WanVideo Model Loader without installing ComfyUI
This node already runs on ComfyICU. Open it in the browser, hit run, and you're billed for GPU seconds — not for setup, and not while you're idle.