Does not work on Nvidia 1080 8GB

Although it is claimed that this program only needs 6 GB of VRAM I get the following error:

Currently enabled native sdp backends: ['flash', 'math', 'mem_efficient', 'cudnn']
Xformers is not installed!
Flash Attn is not installed!
Sage Attn is not installed!
Namespace(share=False, server='127.0.0.1', port=None, inbrowser=True)
Free VRAM 7.0185546875 GB
High-VRAM Mode: False
Downloading shards: 100%|██████████████████████████████████████████████████████████████| 4/4 [00:00<00:00, 4013.69it/s]
Loading checkpoint shards: 100%|█████████████████████████████████████████████████████████| 4/4 [00:00<00:00, 8.51it/s]
Fetching 3 files: 100%|██████████████████████████████████████████████████████████████████████████| 3/3 [00:00

Reproduce the failure from webui/demo_gradio.py, then read the attention path in webui/diffusers_helper/models/hunyuan_video_packed.py and the sampler entry in webui/diffusers_helper/k_diffusion/k_diffusion_hunyuan.py. Compare the reported VRAM state with the scaled dot-product attention call; done means the reported 8 GB configuration completes generation without the shown CUDA out-of-memory error.

Written by the indexing model from the issue text.