Can't start generating on 2060s

Nobody has claimed this yet.

There is 8G VRAM on 2060s and 64G DDR4 on host. I have also left over 200G spare space on the intalled drive.
It will still show as the following:

torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 28.87 GiB. GPU 0 has a total capacity of 8.00 GiB of which 0 bytes is free. Of the allocated memory 31.44 GiB is allocated by PyTorch, and 339.26 MiB 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)
Unloaded DynamicSwap_LlamaModel as complete.
Unloaded CLIPTextModel as complete.
Unloaded SiglipVisionModel as complete.
Unloaded AutoencoderKLHunyuanVideo as complete.
Unloaded DynamicSwap_HunyuanVideoTransformer3DModelPacked as complete.

Start by reproducing the reported CUDA out-of-memory failure on an RTX 2060S with 8 GB VRAM and 64 GB system memory. Review the PyTorch CUDA memory-management documentation and the suggested PYTORCH_CUDA_ALLOC_CONF setting. Done means generation can start on the reported hardware without the shown allocation failure.

Written by the indexing model from the issue text.