[Bug]: RuntimeError: CUDA 11.0 or higher is required to build the package

[Aruen24]

Is there an existing issue ? / 是否已有相关的 issue ?

• I have searched, and there is no existing issue. / 我已经搜索过了，没有相关的 issue。

Describe the bug / 描述这个 bug

pip install inference/vllm

To Reproduce / 如何复现

Processing ./inference/vllm
Installing build dependencies ... done
Getting requirements to build wheel ... error
error: subprocess-exited-with-error

× Getting requirements to build wheel did not run successfully.
│ exit code: 1
╰─> [17 lines of output]
/tmp/pip-build-env-m3ek1eux/overlay/lib/python3.10/site-packages/torch/nn/modules/transformer.py:20: UserWarning: Failed to initialize NumPy: No module named 'numpy' (Triggered internally at ../torch/csrc/utils/tensor_numpy.cpp:84.)
device: torch.device = torch.device(torch._C._get_default_device()), # torch.device('cpu'),
Traceback (most recent call last):
File "/home/wyw/anaconda3/envs/dbgpt_env/lib/python3.10/site-packages/pip/_vendor/pyproject_hooks/_in_process/_in_process.py", line 353, in
main()
File "/home/wyw/anaconda3/envs/dbgpt_env/lib/python3.10/site-packages/pip/_vendor/pyproject_hooks/_in_process/_in_process.py", line 335, in main
json_out['return_val'] = hook(**hook_input['kwargs'])
File "/home/wyw/anaconda3/envs/dbgpt_env/lib/python3.10/site-packages/pip/_vendor/pyproject_hooks/_in_process/_in_process.py", line 118, in get_requires_for_build_wheel
return hook(config_settings)
File "/tmp/pip-build-env-m3ek1eux/overlay/lib/python3.10/site-packages/setuptools/build_meta.py", line 325, in get_requires_for_build_wheel
return self._get_build_requires(config_settings, requirements=['wheel'])
File "/tmp/pip-build-env-m3ek1eux/overlay/lib/python3.10/site-packages/setuptools/build_meta.py", line 295, in _get_build_requires
self.run_setup()
File "/tmp/pip-build-env-m3ek1eux/overlay/lib/python3.10/site-packages/setuptools/build_meta.py", line 311, in run_setup
exec(code, locals())
File "", line 108, in
RuntimeError: CUDA 11.0 or higher is required to build the package.
[end of output]

note: This error originates from a subprocess, and is likely not a problem with pip.
error: subprocess-exited-with-error

× Getting requirements to build wheel did not run successfully.
│ exit code: 1
╰─> See above for output.

note: This error originates from a subprocess, and is likely not a problem with pip.

Expected behavior / 期望的结果

No response

Screenshots / 截图

No response

Environment / 环境

- OS: [e.g. Ubuntu 20.04]
- Pytorch: [e.g. torch 2.0.0]
- CUDA: [e.g. CUDA 11.8]
- Device: [e.g. A10, RTX3090]

Additional context / 其他信息

No response

[Jay7979]

What is bug

[zhou-jingyu]

I got similar issue too. It is possibly due to your system selected a wrong version of nvcc. Can be checked by:

nvcc -V
which nvcc

Modify the PATH value to force your system to find the new version of nvcc may resolve the issue:
export PATH="/usr/local/cuda/bin:$PATH"

[yingwei13mei]

I got the similar error when pip install inference/vllm. As following info:
File "C:\Users\Administrator\AppData\Local\Temp\pip-build-env-lzrf58sq\overlay\Lib\site-packages\setuptools\build_meta.py", line 311, in run_setup
exec(code, locals())
File "", line 30, in
RuntimeError: Cannot find CUDA_HOME. CUDA must be available to build the package.
[end of output]

note: This error originates from a subprocess, and is likely not a problem with pip.
error: subprocess-exited-with-error

× Getting requirements to build wheel did not run successfully.
│ exit code: 1
╰─> See above for output.

note: This error originates from a subprocess, and is likely not a problem with pip.

• OS: [Win 10]
• Pytorch: [2.1.0+cu118]
• CUDA: [CUDA 11.3]
• Device: [RTX3060]

(base) F:\MiniCPM>nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2021 NVIDIA Corporation
Built on Mon_May__3_19:41:42_Pacific_Daylight_Time_2021
Cuda compilation tools, release 11.3, V11.3.109
Build cuda_11.3.r11.3/compiler.29920130_0