May I ask if the gguf inference for wan2.2-i2v is supported?

Start in pipelines/wan_video.py at from_pretrained and _from_state_dict, then inspect models/wan/wan_dit.py at from_state_dict and the failing load_state_dict call. Compare the checkpoint's patch_embedding shape with the model configuration to determine whether Wan2.2-I2V GGUF is supported or the usage is mismatched; done means the supported configuration or required correction is documented.

May I ask if the gguf inference for wan2.2-i2v is supported?
use model QuantStack/Wan2.2-T2V-A14B-GGUF

`model_path = [
"/model_files/Wan2.2-I2V-A14B-GGUF/LowNoise/Wan2.2-I2V-A14B-LowNoise-Q4_K_S.gguf",
"/model_files/Wan2.2-I2V-A14B-GGUF/HighNoise/Wan2.2-I2V-A14B-HighNoise-Q4_K_S.gguf"
]
t5_path = "/model_files/wan2.1-umt5/umt5.safetensors"
vae_path = "/model_files/muse/wan2.2-vae/__version/vae.safetensors"

config = WanPipelineConfig(
model_path=model_path,
t5_path=t5_path,
vae_path=vae_path,
device="cuda:0",
offload_mode="cpu_offload"
)
pipe = WanVideoPipeline.from_pretrained(config)`

It seems that it cannot be operated.
File "/miniconda3/envs/diffSynth-engine/lib/python3.10/site-packages/diffsynth_engine/pipelines/wan_video.py", line 506, in from_pretrained
return cls.from_state_dict(state_dicts, config)
File "/miniconda3/envs/diffSynth-engine/lib/python3.10/site-packages/diffsynth_engine/pipelines/wan_video.py", line 520, in from_state_dict
pipe = cls._from_state_dict(state_dicts, config)
File "/miniconda3/envs/diffSynth-engine/lib/python3.10/site-packages/diffsynth_engine/pipelines/wan_video.py", line 566, in _from_state_dict
dit = WanDiT.from_state_dict(
File "miniconda3/envs/diffSynth-engine/lib/python3.10/site-packages/diffsynth_engine/models/wan/wan_dit.py", line 417, in from_state_dict
model.load_state_dict(state_dict, assign=assign)
File "/miniconda3/envs/diffSynth-engine/lib/python3.10/site-packages/diffsynth_engine/models/base.py", line 20, in load_state_dict
super().load_state_dict(state_dict, strict=strict, assign=assign)
File "/miniconda3/envs/diffSynth-engine/lib/python3.10/site-packages/torch/nn/modules/module.py", line 2581, in load_state_dict
raise RuntimeError(
RuntimeError: Error(s) in loading state_dict for WanDiT:
size mismatch for patch_embedding.weight: copying a param with shape torch.Size([5120, 36, 1, 2, 2]) from checkpoint, the shape in current model is torch.Size([5120, 16, 1, 2, 2])

Is it my usage method that has the problem?
Thank you very much for being able to clarify my doubts.