OSError: vikp/surya_det3 does not appear to have a file named pytorch_model.bin...

[menezes-]

Hi!

After updating to version 1.3.3, when trying to use the library from Python code (as the docs suggest) I keep getting this error:

   File "/python3.13/site-packages/marker/models.py", line 18, in create_model_dict
     "detection_model": DetectionPredictor(device=device, dtype=dtype),
                        ~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^
   File "/python3.13/site-packages/surya/common/predictor.py", line 23, in __init__
     self.model = loader.model(device, dtype)
                  ~~~~~~~~~~~~^^^^^^^^^^^^^^^
   File "/python3.13/site-packages/surya/detection/loader.py", line 33, in model
     model = EfficientViTForSemanticSegmentation.from_pretrained(
         self.checkpoint,
     ...<2 lines>...
         revision=self.revision
     )
   File "/python3.13/site-packages/transformers/modeling_utils.py", line 3880, in from_pretrained
     raise EnvironmentError(
     ...<3 lines>...
     )
 OSError: vikp/surya_det3 does not appear to have a file named pytorch_model.bin, model.safetensors, tf_model.h5, model.ckpt or flax_model.msgpack.

Any suggestions on how to fix this? Thanks!

[menezes-]

This was an issue on my setup