Skywork/SkyReels-V2-DF-14B-720P · Unable to run examples with diffusion forcing models

I created a conda environment setting python to 3.10.12. I downloaded SkyReels-V2 from git hub. I ran "pip install -r requirements.txt". I then tried to run one of the example scripts, with the only thing different being that I set the HF_HOME environment variable to control where the models were downloaded. The following output shows the error.
model_id=Skywork/SkyReels-V2-DF-14B-720P
python3 generate_video_df.py \
  --model_id ${model_id} \
  --resolution 540P \
  --ar_step 5 \
  --causal_block_size 5 \
  --base_num_frames 97 \
  --num_frames 737 \
  --overlap_history 17 \
  --image data/in/Kim-AI-0128.png \
  --prompt "A woman with a black t-shirt and brown hair looks directly at the camera while giving an apologetic speech." \
  --addnoise_condition 20 \
  --offload
/workspace/miniconda3/envs/skyreels/lib/python3.10/site-packages/transformers/utils/hub.py:106: FutureWarning: Using `TRANSFORMERS_CACHE` is deprecated and will be removed in v5 of Transformers. Use `HF_HOME` instead.
  warnings.warn(
models_t5_umt5-xxl-enc-bf16.pth: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 11.4G/11.4G [00:07<00:00, 194MB/s]
(…)pytorch_model-00005-of-00006.safetensors: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 9.84G/9.84G [00:09<00:00, 202MB/s]
(…)pytorch_model-00003-of-00006.safetensors: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 9.94G/9.94G [00:20<00:00, 167MB/s]
Fetching 19 files: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 19/19 [00:22<00:00,  1.16s/it]
model_id: /workspace/SkyReels2-Models/hub/models--Skywork--SkyReels-V2-DF-14B-720P/snapshots/21bae7ad4dd1335500988ce7ad56d6c532fc8a44
/workspace/miniconda3/envs/skyreels/lib/python3.10/site-packages/diffusers/configuration_utils.py:248: FutureWarning: It is deprecated to pass a pretrained model name or path to `from_config`.If you were trying to load a model, please use <class 'skyreels_v2_infer.modules.transformer.WanModel'>.load_config(...) followed by <class 'skyreels_v2_infer.modules.transformer.WanModel'>.from_config(...) instead. Otherwise, please make sure to pass a configuration dictionary instead. This functionality will be removed in v1.0.0.
  deprecate("config-passed-as-path", "1.0.0", deprecation_message, standard_warn=False)
Traceback (most recent call last):
  File "/workspace/SkyReels-V2/generate_video_df.py", line 130, in <module>
    pipe = DiffusionForcingPipeline(
  File "/workspace/SkyReels-V2/skyreels_v2_infer/pipelines/diffusion_forcing_pipeline.py", line 55, in __init__
    self.transformer = get_transformer(dit_path, load_device, weight_dtype)
  File "/workspace/SkyReels-V2/skyreels_v2_infer/modules/__init__.py", line 37, in get_transformer
    state_dict = load_file(file_path)
  File "/workspace/miniconda3/envs/skyreels/lib/python3.10/site-packages/safetensors/torch.py", line 313, in load_file
    with safe_open(filename, framework="pt", device=device) as f:
safetensors_rust.SafetensorError: Error while deserializing header: MetadataIncompleteBuffer