Inference time model loading issue

[krishnakanthnakka]

Hello authors,

Can you please advise if I could load the model using AutoModel function similar to one used for InternVL2.5 like

model = AutoModel.from_pretrained(
    model_name,
    torch_dtype=torch.bfloat16,
    low_cpu_mem_usage=True,
    use_flash_attn=True,
    trust_remote_code=True).eval().cuda()

It seems to throw an error about checkpoint file missing.

warnings.warn(f"Importing from {__name__} is deprecated, please import via timm.layers", FutureWarning)
Downloading shards:   0%|                               | 0/7 [00:00<?, ?it/s]
Traceback (most recent call last):
File "/workspace/CyclingQA/lib/models/spatialreasoner_r1.py", line 103, in <module>
  model = AutoModel.from_pretrained(
          ^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/transformers/models/auto/auto_factory.py", line 559, in from_pretrained
  return model_class.from_pretrained(
         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/transformers/modeling_utils.py", line 3715, in from_pretrained
  resolved_archive_file, sharded_metadata = get_checkpoint_shard_files(
                                            ^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/transformers/utils/hub.py", line 1079, in get_checkpoint_shard_files
  cached_filename = cached_file(
                    ^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/transformers/utils/hub.py", line 397, in cached_file
  raise EnvironmentError(f"Could not locate {full_filename} inside {path_or_repo_id}.")
OSError: Could not locate model-00001-of-00007.safetensors inside PLAN-Lab/SpatialReasoner-R1.