NVIDIA
missing tensorrt_bindings for tensorrt 10.0.1
Description
I am trying to pip install tensorrt 10.0.1 but it reports ''ModuleNotFoundError: No module named 'tensorrt_bindings'''
pip install tensorrt==10.0.1 then import tensorrt
In https://pypi.nvidia.com/tensorrt-bindings, there is no tensorrt_bindings==10.0.1.
Is it expected?
Environment
TensorRT Version: 10.0.1
NVIDIA GPU:
NVIDIA Driver Version:
CUDA Version: 12.1.1
CUDNN Version:
Operating System:
Python Version (if applicable):
Tensorflow Version (if applicable):
PyTorch Version (if applicable):
Baremetal or Container (if so, version):
Relevant Files
Model link:
Steps To Reproduce
Commands or scripts:
Have you tried the latest release?:
Can this model run on other frameworks? For example run ONNX model with ONNXRuntime (polygraphy run <model.onnx> --onnxrt):
fixed by pip install tensorrt_cu12_libs==10.0.1 tensorrt_cu12_bindings==10.0.1 tensorrt==10.0.1 --extra-index-url https://pypi.nvidia.com
we shouldn't close this issue.
pip install tensorrt should take care of dependencies properly. the user should not be expected to manually install these other libraries.
What is the status on this issue please? This does not appear to be solved with Tensorrt version 10.6.0.
Note that pip install tensorrt-bindings does not seem to work, one must explicitly install tensorrt-cu12-bindings.
Using uv:
uv add --upgrade torch>=2.6.0 torchaudio>=2.6.0 torchvision>=0.21.0 tensorrt-cu12 tensorrt-cu12-bindings tensorrt-cu12-libs
I end up with the following dependencies:
[project]
dependencies = [
"tensorrt-cu12>=10.8.0.43",
"tensorrt-cu12-bindings>=10.8.0.43",
"tensorrt-cu12-libs>=10.8.0.43",
"torch>=2.6.0",
"torchaudio>=2.6.0",
"torchvision>=0.21.0",
]
This seems to work without issue. I assume pip will be the same
