Error with torch when trying to generate text to video

What happened?
I get an error whenever I try to generate. I currently have Hunyuan Video 1.5, Text2Video 8B, 720p, selected

Your system (OS / GPU / RAM / VRAM / etc.)
Windows 10, AMD RX 6700 XT

Logs / full error output

Pinning data of 'transformer' to reserved RAM
The whole model was pinned to reserved RAM: 33 large blocks spread across 8120.37 MB
Hooked to model 'transformer' (HYVideoDiffusionTransformer)
Async loading plan for model 'transformer' : base size of 14.40 MB will be preloaded with a 144.24 MB async shuttle
Hooked to model 'text_encoder' (Qwen2_5_VLForConditionalGeneration)
Async loading plan for model 'text_encoder' : base size of 2081.88 MB will be preloaded with a 222.37 MB async shuttle
Hooked to model 'vae' (AutoencoderKLConv3D)
Hooked to model 'byt5_model' (T5Stack)
Traceback (most recent call last):
  File "G:\Pinokio\api\wan2gp-amd.git\app\wgp.py", line 7920, in generate_media
    samples = wan_model.generate(
              ^^^^^^^^^^^^^^^^^^^
  File "G:\Pinokio\api\wan2gp-amd.git\app\models\hyvideo\hunyuan.py", line 1031, in generate
    samples = self.pipeline(
              ^^^^^^^^^^^^^^
  File "G:\Pinokio\api\wan2gp-amd.git\app\env\Lib\site-packages\torch\utils\_contextlib.py", line 124, in decorate_context
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "G:\Pinokio\api\wan2gp-amd.git\app\models\hyvideo\diffusion\pipelines\pipeline_hunyuan_video.py", line 1345, in __call__
    device = torch.device(f"cuda:{dist.get_rank()}") if dist.is_initialized() else self._execution_device
                                                        ^^^^^^^^^^^^^^^^^^^
AttributeError: module 'torch.distributed' has no attribute 'is_initialized'
Error Queue autosaved successfully to error_queue.zip