theluyuan/DiffSynth-Studio
1
0
Fork 0
mirror of https://github.com/modelscope/DiffSynth-Studio.git synced 2026-03-18 22:08:13 +00:00
Code Issues Packages Projects Releases Wiki Activity

Fix dtype issue in time embedding calculation

Browse Source
This commit is contained in:
Yudong Jin
2025-11-01 03:11:03 +08:00
committed by GitHub
parent a30ed9093f
commit b8d7d55568
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
diffsynth/models/wan_video_dit.py
View File

@@ -362,7 +362,7 @@ class WanModel(torch.nn.Module):
**kwargs,
):
t = self.time_embedding(
sinusoidal_embedding_1d(self.freq_dim, timestep))
sinusoidal_embedding_1d(self.freq_dim, timestep).to(x.dtype))
t_mod = self.time_projection(t).unflatten(1, (6, self.dim))
context = self.text_embedding(context)