WSL 2 -> AttributeError: 'DINOv3ViTModel' object has no attribute 'layer'

I'm using Ubuntu on WSL 2 under Windows 11, managed to pass the gated models issue but now I hit this error when running the example.py file:

~/TRELLIS.2$ python example.py
[SPARSE] Conv backend: flex_gemm; Attention backend: flash_attn
[ATTENTION] Using backend: flash_attn
Loading weights: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████| 415/415 [00:00<00:00, 3895.83it/s]
Loading weights: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████| 754/754 [00:00<00:00, 3714.00it/s]
Traceback (most recent call last):
  File "/home/vl4dt/TRELLIS.2/example.py", line 25, in <module>
    mesh = pipeline.run(image)[0]
  File "/home/vl4dt/miniforge3/envs/trellis2/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 116, in decorate_context
    return func(*args, **kwargs)
  File "/home/vl4dt/TRELLIS.2/trellis2/pipelines/trellis2_image_to_3d.py", line 539, in run
    cond_512 = self.get_cond([image], 512)
  File "/home/vl4dt/TRELLIS.2/trellis2/pipelines/trellis2_image_to_3d.py", line 177, in get_cond
    cond = self.image_cond_model(image)
  File "/home/vl4dt/miniforge3/envs/trellis2/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 116, in decorate_context
    return func(*args, **kwargs)
  File "/home/vl4dt/TRELLIS.2/trellis2/modules/image_feature_extractor.py", line 117, in __call__
    features = self.extract_features(image)
  File "/home/vl4dt/TRELLIS.2/trellis2/modules/image_feature_extractor.py", line 86, in extract_features
    for i, layer_module in enumerate(self.model.layer):
  File "/home/vl4dt/miniforge3/envs/trellis2/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1928, in __getattr__
    raise AttributeError(
AttributeError: 'DINOv3ViTModel' object has no attribute 'layer'

Tried downgraded the transformers library but ran into other issues, I'm at a loss here.

pip-freeze.txt

Reproduce the failure with python example.py, then inspect trellis2/modules/image_feature_extractor.py, especially extract_features, and the call path in trellis2/pipelines/trellis2_image_to_3d.py. Compare the loaded DINOv3ViTModel with the access shown in the traceback and verify that example.py completes without this AttributeError.