diff --git a/nodes.py b/nodes.py
index f42c435..165d95c 100644
--- a/nodes.py
+++ b/nodes.py
@@ -4361,11 +4361,10 @@ class WanVideoSampler:
 
                                 vae.model.clear_cache()
                                 vae.to(offload_device)
-                                mm.soft_empty_cache()
-                                gc.collect()
                                 
                                 temporal_ref_latents = torch.cat([msk, temporal_ref_latents], dim=0) # 4+C T H W
                                 image_cond_in = torch.cat([ref_latent.to(device), temporal_ref_latents], dim=1) # 4+C T+trefs H W
+                                del temporal_ref_latents, msk, bg_image_slice
 
                             noise = torch.randn(16, latent_window_size + 1, lat_h, lat_w, dtype=torch.float32, device=torch.device("cpu"), generator=seed_g).to(device)
                             seq_len = math.ceil((noise.shape[2] * noise.shape[3]) / 4 * noise.shape[1])
@@ -4379,6 +4378,7 @@ class WanVideoSampler:
                                     pad_len = latent_window_size - pose_input_slice.shape[2]
                                     pad = torch.zeros(pose_input_slice.shape[0], pose_input_slice.shape[1], pad_len, pose_input_slice.shape[3], pose_input_slice.shape[4], device=pose_input_slice.device, dtype=pose_input_slice.dtype)
                                     pose_input_slice = torch.cat([pose_input_slice, pad], dim=2)
+                                    del pad
                                 pose_input_slice = pose_input_slice.to(device, dtype)
                             
                             if samples is not None:
@@ -4496,7 +4496,6 @@ class WanVideoSampler:
                             videos = vae.decode(latent[:, 1:].unsqueeze(0).to(device, vae.dtype), device=device, tiled=tiled_vae, pbar=False)[0].cpu()
                             del latent
                             vae.model.clear_cache()
-                            vae.to(offload_device)
 
                             sampling_pbar.close()
                             
@@ -4540,6 +4539,7 @@ class WanVideoSampler:
                             gen_video_samples = torch.zeros(3, 1, 64, 64) # dummy output
 
                         if force_offload:
+                            vae.to(offload_device)
                             if not model["auto_cpu_offload"]:
                                 offload_transformer(transformer)
                         try: