mirrors/ComfyUI
1
0
Fork 0
mirror of https://github.com/comfyanonymous/ComfyUI.git synced 2025-09-10 03:25:22 +00:00
Code Issues Packages Projects Releases Wiki Activity

Support wan2.2 5B fun control model. (#9611)

Browse Source 
Use the Wan22FunControlToVideo node.
This commit is contained in:
comfyanonymous
2025-08-28 19:13:07 -07:00
committed by GitHub
parent d28b39d93d
commit e80a14ad50
2 changed files with 18 additions and 16 deletions
Download Patch File Download Diff File Expand all files Collapse all files

19
comfy_extras/nodes_wan.py
View File

@@ -139,16 +139,21 @@ class Wan22FunControlToVideo(io.ComfyNode):
@classmethod
def execute(cls, positive, negative, vae, width, height, length, batch_size, ref_image=None, start_image=None, control_video=None) -> io.NodeOutput:
latent = torch.zeros([batch_size, 16, ((length - 1) // 4) + 1, height // 8, width // 8], device=comfy.model_management.intermediate_device())
concat_latent = torch.zeros([batch_size, 16, ((length - 1) // 4) + 1, height // 8, width // 8], device=comfy.model_management.intermediate_device())
concat_latent = comfy.latent_formats.Wan21().process_out(concat_latent)
spacial_scale = vae.spacial_compression_encode()
latent_channels = vae.latent_channels
latent = torch.zeros([batch_size, latent_channels, ((length - 1) // 4) + 1, height // spacial_scale, width // spacial_scale], device=comfy.model_management.intermediate_device())
concat_latent = torch.zeros([batch_size, latent_channels, ((length - 1) // 4) + 1, height // spacial_scale, width // spacial_scale], device=comfy.model_management.intermediate_device())
if latent_channels == 48:
concat_latent = comfy.latent_formats.Wan22().process_out(concat_latent)
else:
concat_latent = comfy.latent_formats.Wan21().process_out(concat_latent)
concat_latent = concat_latent.repeat(1, 2, 1, 1, 1)
mask = torch.ones((1, 1, latent.shape[2] * 4, latent.shape[-2], latent.shape[-1]))
if start_image is not None:
start_image = comfy.utils.common_upscale(start_image[:length].movedim(-1, 1), width, height, "bilinear", "center").movedim(1, -1)
concat_latent_image = vae.encode(start_image[:, :, :, :3])
concat_latent[:,16:,:concat_latent_image.shape[2]] = concat_latent_image[:,:,:concat_latent.shape[2]]
concat_latent[:,latent_channels:,:concat_latent_image.shape[2]] = concat_latent_image[:,:,:concat_latent.shape[2]]
mask[:, :, :start_image.shape[0] + 3] = 0.0
ref_latent = None
@@ -159,11 +164,11 @@ class Wan22FunControlToVideo(io.ComfyNode):
if control_video is not None:
control_video = comfy.utils.common_upscale(control_video[:length].movedim(-1, 1), width, height, "bilinear", "center").movedim(1, -1)
concat_latent_image = vae.encode(control_video[:, :, :, :3])
concat_latent[:,:16,:concat_latent_image.shape[2]] = concat_latent_image[:,:,:concat_latent.shape[2]]
concat_latent[:,:latent_channels,:concat_latent_image.shape[2]] = concat_latent_image[:,:,:concat_latent.shape[2]]
mask = mask.view(1, mask.shape[2] // 4, 4, mask.shape[3], mask.shape[4]).transpose(1, 2)
positive = node_helpers.conditioning_set_values(positive, {"concat_latent_image": concat_latent, "concat_mask": mask, "concat_mask_index": 16})
negative = node_helpers.conditioning_set_values(negative, {"concat_latent_image": concat_latent, "concat_mask": mask, "concat_mask_index": 16})
positive = node_helpers.conditioning_set_values(positive, {"concat_latent_image": concat_latent, "concat_mask": mask, "concat_mask_index": latent_channels})
negative = node_helpers.conditioning_set_values(negative, {"concat_latent_image": concat_latent, "concat_mask": mask, "concat_mask_index": latent_channels})
if ref_latent is not None:
positive = node_helpers.conditioning_set_values(positive, {"reference_latents": [ref_latent]}, append=True)