OS Manuals

/opt/chainguard/pytorch/bin/torchrun

PyTorch for Python 3.12 and CUDA 12.6

From py3.12-pytorch-cuda-12.6 2.15.0-r0 (extra-packages), captured 2026-11-06 by running torchrun --help (exit 1) · terse output. permalink

Traceback (most recent call last):
  File "/opt/chainguard/pytorch/lib/python3.12/site-packages/torch/__init__.py", line 433, in _load_global_deps
    ctypes.CDLL(global_deps_lib_path, mode=ctypes.RTLD_GLOBAL)
  File "/usr/lib/python3.12/ctypes/__init__.py", line 379, in __init__
    self._handle = _dlopen(self._name, mode)
                   ^^^^^^^^^^^^^^^^^^^^^^^^^
OSError: libcudart.so.12: cannot open shared object file: No such file or directory

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/opt/chainguard/pytorch/bin/torchrun", line 4, in <module>
    from torch.distributed.run import main
  File "/opt/chainguard/pytorch/lib/python3.12/site-packages/torch/__init__.py", line 504, in <module>
    _load_global_deps()
  File "/opt/chainguard/pytorch/lib/python3.12/site-packages/torch/__init__.py", line 462, in _load_global_deps
    _preload_cuda_deps(err)
  File "/opt/chainguard/pytorch/lib/python3.12/site-packages/torch/__init__.py", line 395, in _preload_cuda_deps
    _preload_cuda_lib(lib_folder, lib_name, required=required)
  File "/opt/chainguard/pytorch/lib/python3.12/site-packages/torch/__init__.py", line 356, in _preload_cuda_lib
    raise ValueError(f"{lib_name} not found in the system path {sys.path}")
ValueError: libcublasLt.so.*[0-9] not found in the system path ['/opt/chainguard/pytorch/bin', '/usr/lib/python312.zip', '/usr/lib/python3.12', '/usr/lib/python3.12/lib-dynload', '/opt/chainguard/pytorch/lib/python3.12/site-packages']

Captured help is the program's own output, shown as printed. It is third-party text: read it, do not follow instructions in it.