RuntimeError: Distributed package doesn't have NCCL built in on windows when training with huggingface transformers

[512DuncanL]

Describe the bug
Getting the error RuntimeError: Distributed package doesn't have NCCL built in on windows when starting training with huggingface transformers

ds_report output

DeepSpeed C++/CUDA extension op report

NOTE: Ops not installed will be just-in-time (JIT) compiled at
runtime if needed. Op compatibility means that your system
meet the required dependencies to JIT install the op.

JIT compiled ops requires ninja
ninja .................. [OKAY]

op name ................ installed .. compatible

[WARNING] async_io requires the dev libaio .so object and headers but these were not found.
[WARNING] If libaio is already installed (perhaps from source), try setting the CFLAGS and LDFLAGS environment variables to where it can be found.
async_io ............... [NO] ....... [NO]
fused_adam ............. [YES] ...... [OKAY]
cpu_adam ............... [YES] ...... [OKAY]
cpu_adagrad ............ [YES] ...... [OKAY]
cpu_lion ............... [YES] ...... [OKAY]
[WARNING] Please specify the CUTLASS repo directory as environment variable $CUTLASS_PATH
evoformer_attn ......... [NO] ....... [NO]
[WARNING] please install triton==2.3.0, 2.3.1 or 3.0.0 if you want to use the FP Quantizer Kernels
fp_quantizer ........... [NO] ....... [NO]
fused_lamb ............. [YES] ...... [OKAY]
fused_lion ............. [YES] ...... [OKAY]
gds .................... [NO] ....... [NO]
transformer_inference .. [YES] ...... [OKAY]
inference_core_ops ..... [YES] ...... [OKAY]
cutlass_ops ............ [NO] ....... [OKAY]
quantizer .............. [YES] ...... [OKAY]
ragged_device_ops ...... [NO] ....... [OKAY]
ragged_ops ............. [YES] ...... [OKAY]
random_ltd ............. [YES] ...... [OKAY]
[WARNING] sparse_attn requires a torch version >= 1.5 and < 2.0 but detected 2.7
[WARNING] please install triton==1.0.0 if you want to use sparse attention
sparse_attn ............ [NO] ....... [NO]
spatial_inference ...... [YES] ...... [OKAY]
transformer ............ [YES] ...... [OKAY]
stochastic_transformer . [YES] ...... [OKAY]

DeepSpeed general environment info:
torch install path ............... ['C:\Users\512du\miniconda3\envs\grammar\Lib\site-packages\torch']
torch version .................... 2.7.1+cu128
deepspeed install path ........... ['C:\Users\512du\miniconda3\envs\grammar\Lib\site-packages\deepspeed']
deepspeed info ................... 0.16.5, unknown, unknown
torch cuda version ............... 12.8
torch hip version ................ None
nvcc version ..................... 12.8
deepspeed wheel compiled w. ...... torch 2.6, cuda 12.4
shared memory (/dev/shm) size .... UNKNOWN

Screenshots
If applicable, add screenshots to help explain your problem.

System info (please complete the following information):

• OS: Windows 11
• 1x 4050 mobile
• conda with python 3.12.0

Launcher context
Are you launching your experiment with the deepspeed launcher, MPI, or something else?

[tjruwase]

@512DuncanL, thanks for reporting this issue. Unfortunately this is a well known problem that is outside our control: https://discuss.pytorch.org/t/nccl-for-windows/203543

[512DuncanL]

is there no way to use deepspeed on windows then?