Failed sanity test for GPU install v2.7.10+xpu (Windows/pip) with ModuleNotFoundError: No module named 'oneccl_bindings_for_pytorch'

[rkilchmn]

Describe the bug

I installed GPU v2.7.10+xpu (Windows/pip) following the instructions.

Then I ran the sanity test:
python -c "import torch; import intel_extension_for_pytorch as ipex; print(torch.__version__); print(ipex.__version__); [print(f'[{i}]: {torch.xpu.get_device_properties(i)}') for i in range(torch.xpu.device_count())];"

and go this import error: ModuleNotFoundError: No module named 'oneccl_bindings_for_pytorch'

python -c "import torch; import intel_extension_for_pytorch as ipex; print(torch.__version__); print(ipex.__version__); [print(f'[{i}]: {torch.xpu.get_device_properties(i)}') for i in range(torch.xpu.device_count())];" [W430 12:16:51.000000000 OperatorEntry.cpp:161] Warning: Warning only once for all operators, other operators may also be overridden. Overriding a previously registered kernel for the same operator and the same dispatch key operator: aten::geometric_(Tensor(a!) self, float p, *, Generator? generator=None) -> Tensor(a!) registered at C:\actions-runner\_work\pytorch\pytorch\pytorch\build\aten\src\ATen\RegisterSchema.cpp:6 dispatch key: XPU previous kernel: registered at C:\actions-runner\_work\pytorch\pytorch\pytorch\aten\src\ATen\VmapModeRegistrations.cpp:37 new kernel: registered at H:\frameworks.ai.pytorch.ipex-gpu\build\Release\csrc\gpu\csrc\gpu\xpu\ATen\RegisterXPU_0.cpp:186 (function operator ()) Traceback (most recent call last): File "<string>", line 1, in <module> File "c:\Users\myuser\Documents\project\openedai-whisper-ipex-llm\.conda\Lib\site-packages\intel_extension_for_pytorch\__init__.py", line 127, in <module> from . import xpu File "c:\Users\myuser\Documents\project\openedai-whisper-ipex-llm\.conda\Lib\site-packages\intel_extension_for_pytorch\xpu\__init__.py", line 20, in <module> from .utils import * File "c:\Users\myuser\Documents\project\openedai-whisper-ipex-llm\.conda\Lib\site-packages\intel_extension_for_pytorch\xpu\utils.py", line 6, in <module> from .. import frontend File "c:\Users\myuser\Documents\project\openedai-whisper-ipex-llm\.conda\Lib\site-packages\intel_extension_for_pytorch\frontend.py", line 9, in <module> from .nn import utils File "c:\Users\myuser\Documents\project\openedai-whisper-ipex-llm\.conda\Lib\site-packages\intel_extension_for_pytorch\nn\utils\__init__.py", line 1, in <module> from intel_extension_for_pytorch.nn.utils import _weight_prepack File "c:\Users\myuser\Documents\project\openedai-whisper-ipex-llm\.conda\Lib\site-packages\intel_extension_for_pytorch\nn\utils\_weight_prepack.py", line 121, in <module> from deepspeed import comm File "c:\Users\myuser\Documents\project\openedai-whisper-ipex-llm\.conda\Lib\site-packages\deepspeed\__init__.py", line 25, in <module> from . import ops File "c:\Users\myuser\Documents\project\openedai-whisper-ipex-llm\.conda\Lib\site-packages\deepspeed\ops\__init__.py", line 6, in <module> from . import adam File "c:\Users\myuser\Documents\project\openedai-whisper-ipex-llm\.conda\Lib\site-packages\deepspeed\ops\adam\__init__.py", line 6, in <module> from .cpu_adam import DeepSpeedCPUAdam File "c:\Users\myuser\Documents\project\openedai-whisper-ipex-llm\.conda\Lib\site-packages\deepspeed\ops\adam\cpu_adam.py", line 8, in <module> from deepspeed.utils import logger File "c:\Users\myuser\Documents\project\openedai-whisper-ipex-llm\.conda\Lib\site-packages\deepspeed\utils\__init__.py", line 10, in <module> from .groups import * File "c:\Users\myuser\Documents\project\openedai-whisper-ipex-llm\.conda\Lib\site-packages\deepspeed\utils\groups.py", line 28, in <module> from deepspeed import comm as dist File "c:\Users\myuser\Documents\project\openedai-whisper-ipex-llm\.conda\Lib\site-packages\deepspeed\comm\__init__.py", line 7, in <module> from .comm import * File "c:\Users\myuser\Documents\project\openedai-whisper-ipex-llm\.conda\Lib\site-packages\deepspeed\comm\comm.py", line 31, in <module> from deepspeed.comm.ccl import CCLBackend File "c:\Users\myuser\Documents\project\openedai-whisper-ipex-llm\.conda\Lib\site-packages\deepspeed\comm\ccl.py", line 11, in <module> from deepspeed.ops.op_builder import NotImplementedBuilder File "c:\Users\myuser\Documents\project\openedai-whisper-ipex-llm\.conda\Lib\site-packages\deepspeed\ops\op_builder\__init__.py", line 53, in <module> this_module.__dict__[member_name] = builder_closure(member_name) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "c:\Users\myuser\Documents\project\openedai-whisper-ipex-llm\.conda\Lib\site-packages\deepspeed\ops\op_builder\__init__.py", line 41, in builder_closure builder = get_accelerator().get_op_builder(member_name) ^^^^^^^^^^^^^^^^^ File "c:\Users\myuser\Documents\project\openedai-whisper-ipex-llm\.conda\Lib\site-packages\deepspeed\accelerator\real_accelerator.py", line 186, in get_accelerator from .xpu_accelerator import XPU_Accelerator File "c:\Users\myuser\Documents\project\openedai-whisper-ipex-llm\.conda\Lib\site-packages\deepspeed\accelerator\xpu_accelerator.py", line 9, in <module> **import oneccl_bindings_for_pytorch** # noqa: F401 # type: ignore ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ **ModuleNotFoundError: No module named 'oneccl_bindings_for_pytorch'**

Versions

PyTorch version: 2.7.0+xpu
PyTorch CXX11 ABI: No
IPEX version: N/A
IPEX commit: N/A
Build type: N/A

OS: Microsoft Windows 11 Pro (10.0.26100 64-bit)
GCC version: N/A
Clang version: N/A
IGC version: N/A
CMake version: N/A
Libc version: N/A

Python version: 3.11.11 | packaged by Anaconda, Inc. | (main, Dec 11 2024, 16:34:19) [MSC v.1929 64 bit (AMD64)] (64-bit runtime)
Python platform: Windows-10-10.0.26100-SP0
Is XPU available: N/A
DPCPP runtime: N/A
MKL version: N/A

GPU models and configuration onboard:

• Intel(R) Iris(R) Xe Graphics

GPU models and configuration detected:
N/A

Driver version:

• 32.0.101.5972 (20240819000000.***+)

CPU:
Description: Intel64 Family 6 Model 140 Stepping 1
Manufacturer: GenuineIntel
Name: 11th Gen Intel(R) Core(TM) i5-1135G7 @ 2.40GHz
NumberOfCores: 4
NumberOfEnabledCore: 4
NumberOfLogicalProcessors: 8
ThreadCount: 8

Versions of relevant libraries:
[conda] deepspeed 0.15.0 pypi_0 pypi
[conda] dpcpp-cpp-rt 2025.0.5 pypi_0 pypi
[conda] intel-cmplr-lib-rt 2025.0.5 pypi_0 pypi
[conda] intel-cmplr-lib-ur 2025.0.5 pypi_0 pypi
[conda] intel-cmplr-lic-rt 2025.0.5 pypi_0 pypi
[conda] intel-extension-for-pytorch 2.7.10+xpu pypi_0 pypi
[conda] intel-extension-for-transformers 1.4.2 pypi_0 pypi
[conda] intel-opencl-rt 2025.0.5 pypi_0 pypi
[conda] intel-openmp 2025.0.5 pypi_0 pypi
[conda] intel-pti 0.10.1 pypi_0 pypi
[conda] intel-sycl-rt 2025.0.5 pypi_0 pypi
[conda] libuv 1.48.0 h827c3e9_0
[conda] mkl 2025.0.1 pypi_0 pypi
[conda] mkl-dpcpp 2025.0.1 pypi_0 pypi
[conda] numpy 1.26.4 pypi_0 pypi
[conda] onemkl-sycl-blas 2025.0.1 pypi_0 pypi
[conda] onemkl-sycl-datafitting 2025.0.1 pypi_0 pypi
[conda] onemkl-sycl-dft 2025.0.1 pypi_0 pypi
[conda] onemkl-sycl-lapack 2025.0.1 pypi_0 pypi
[conda] onemkl-sycl-rng 2025.0.1 pypi_0 pypi
[conda] onemkl-sycl-sparse 2025.0.1 pypi_0 pypi
[conda] onemkl-sycl-stats 2025.0.1 pypi_0 pypi
[conda] onemkl-sycl-vm 2025.0.1 pypi_0 pypi
[conda] pytorch-triton-xpu 3.3.0 pypi_0 pypi
[conda] torch 2.7.0+xpu pypi_0 pypi
[conda] torchaudio 2.7.0+xpu pypi_0 pypi
[conda] torchvision 0.22.0+xpu pypi_0 pypi
[conda] transformers 4.51.3 pypi_0 pypi

[rkilchmn]

In the code there is this additional linter instruction:

import oneccl_bindings_for_pytorch # noqa: F401 # type: ignore
Meaning: Ignore F401, which is: "module imported but unused"

Why is it included if not used?

[ZhaoqiongZ]

Hi @rkilchmn, thank you for trying this out. We don't support torch-ccl on Windows. It appears there is deepseek in your environment and deepSeek depends on torch-ccl.
You can have a try with OS Ubuntu or uninstall deepspeed