How to use torch.float8_e4m3fn and lora together?

Hi, I'm trying this Lora (https://huggingface.co/Remade-AI/Squish) with examples/wanvideo/wan_14b_image_to_video.py script and it works great with default dtype (bfloat16), codes like below

model_manager.load_models(
    [
        [
            "models/Wan-AI/Wan2.1-I2V-14B-480P/diffusion_pytorch_model-00001-of-00007.safetensors",
            "models/Wan-AI/Wan2.1-I2V-14B-480P/diffusion_pytorch_model-00002-of-00007.safetensors",
            "models/Wan-AI/Wan2.1-I2V-14B-480P/diffusion_pytorch_model-00003-of-00007.safetensors",
            "models/Wan-AI/Wan2.1-I2V-14B-480P/diffusion_pytorch_model-00004-of-00007.safetensors",
            "models/Wan-AI/Wan2.1-I2V-14B-480P/diffusion_pytorch_model-00005-of-00007.safetensors",
            "models/Wan-AI/Wan2.1-I2V-14B-480P/diffusion_pytorch_model-00006-of-00007.safetensors",
            "models/Wan-AI/Wan2.1-I2V-14B-480P/diffusion_pytorch_model-00007-of-00007.safetensors",
            
            
        ],
        "models/Wan-AI/Wan2.1-I2V-14B-480P/models_t5_umt5-xxl-enc-bf16.pth",
        "models/Wan-AI/Wan2.1-I2V-14B-480P/Wan2.1_VAE.pth",
    ],
    torch_dtype=torch.bfloat16 # You can set `torch_dtype=torch.float8_e4m3fn` to enable FP8 quantization.
)

However, if I just updated torch_dtype=torch.bfloat16 to torch_dtype=torch.float8_e4m3fn without touching any other codes, the lora effect is gone (but inference step can still finish without errors messages)

Kindly request some guidances here on how to use Quantization and Lora together here.

Thanks.

Start with examples/wanvideo/wan_14b_image_to_video.py and the model_manager.load_models call shown in the report. Reproduce the comparison between torch.bfloat16 and torch.float8_e4m3fn with the referenced LoRA, then trace how quantization and LoRA loading interact. Done means documenting or correcting the configuration so the LoRA effect remains observable under FP8.