nodes/kijai-ComfyUI-WanVideoWrapper
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Revert this as it didn't end up working

Browse Source 
Pusa extra latents seem to need to be encoded individually
This commit is contained in:
kijai
2025-09-03 15:48:14 +03:00
parent 54391300b1
commit 493106b555
1 changed files with 2 additions and 15 deletions
Download Patch File Download Diff File Expand all files Collapse all files
nodes.py
+2 -15
View File
@@ -1887,9 +1887,7 @@ class WanVideoSampler:
#I2V
image_cond = image_embeds.get("image_embeds", None)
if image_cond is not None:
if is_pusa:
image_cond_mask = image_embeds.get("mask", None)
elif transformer.in_dim == 16:
if transformer.in_dim == 16:
raise ValueError("T2V (text to video) model detected, encoded images only work with I2V (Image to video) models")
elif transformer.in_dim not in [48, 32]: # fun 2.1 models don't use the mask
image_cond_mask = image_embeds.get("mask", None)
@@ -2315,18 +2313,7 @@ class WanVideoSampler:
# extra latents (Pusa) and 5b
latents_to_insert = add_index = None
extra_latents = image_embeds.get("extra_latents", None)
if extra_latents is None:
if image_cond is not None and is_pusa: # get images for pusa if I2V node is used
extra_latents = image_cond
# Find indices where mask is 1
all_indices = torch.where(image_cond_mask[:, :, 0, 0].any(dim=0))[0].tolist()
num_extra_frames = len(all_indices)
if start_step == 0:
for idx in all_indices:
noise[:, idx] = extra_latents[:, idx].to(noise)
log.info(f"Adding extra sample to latent index {idx}")
image_cond = None
elif extra_latents is not None and transformer.multitalk_model_type.lower() != "infinitetalk":
if extra_latents is not None and transformer.multitalk_model_type.lower() != "infinitetalk":
all_indices = []
for entry in extra_latents:
add_index = entry["index"]