Is 4090 video memory not enough for face swapping?

[ak3389]

python infer_face_swap.py
--prompt "The video features a woman standing in front of a large screen displaying the words ""Tech Minute"" and the logo for CNET. She is wearing a purple top and appears to be presenting or speaking about technology-related topics. The background includes a cityscape with tall buildings, suggesting an urban setting. The woman seems to be engaged in a discussion or providing information on technology news or trends. The overall atmosphere is professional and informative, likely aimed at educating viewers about the latest developments in the tech industry."
--ip_image "test/input/ruonan.jpg"
--output "test/output/ruonan.mp4"
--denoising_strength 0.85

Traceback (most recent call last):
File "/root/lanyun-tmp/Stand-In/infer_face_swap.py", line 104, in
video = pipe(
^^^^^
File "/root/miniconda/envs/Stand-In/lib/python3.11/site-packages/torch/utils/_contextlib.py", line 116, in decorate_context
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/root/lanyun-tmp/Stand-In/pipelines/wan_video_face_swap.py", line 598, in call
noise_pred_posi = self.model_fn(
^^^^^^^^^^^^^^
File "/root/lanyun-tmp/Stand-In/pipelines/wan_video_face_swap.py", line 1762, in model_fn_wan_video
x, x_ip = block(x, context, t_mod, freqs, x_ip=x_ip, t_mod_ip=t_mod_ip)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/root/miniconda/envs/Stand-In/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1751, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/root/miniconda/envs/Stand-In/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1762, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/root/lanyun-tmp/Stand-In/models/wan_video_dit.py", line 361, in forward
attn_out = self.self_attn(input_x, freqs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/root/miniconda/envs/Stand-In/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1751, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/root/miniconda/envs/Stand-In/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1762, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/root/lanyun-tmp/Stand-In/models/wan_video_dit.py", line 214, in forward
q_main = rope_apply(q_main, freqs_main, self.num_heads)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/root/lanyun-tmp/Stand-In/models/wan_video_dit.py", line 116, in rope_apply
x_out = torch.view_as_real(x_out * freqs).flatten(2)
~~~~~~^~~~~~~
torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 1.25 GiB. GPU 0 has a total capacity of 23.55 GiB of which 388.69 MiB is free. Including non-PyTorch memory, this process has 0 bytes memory in use. Of the allocated memory 16.91 GiB is allocated by PyTorch, and 3.77 GiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)

[KBRASK]

You can use Kijai's quantization workflow, which is faster and uses less GPU memory.

[rzgarespo]

same problem with the wan2.2 using kija wvr (fp8, 36GB VRAM)

update:
changing the 512x848 to 480x720 solved the issue in my end.