No module named onnx_graphsurgeon (StableDiffusion WebUI - TensorRT extension)

AI & Data Science Deep Learning (Training & Inference) TensorRT
1

I’m utilizing the Stable Diffusion WebUI package installed via the Stability Matrix installer. To integrate TensorRT functionality, I accessed GitHub - NVIDIA/TensorRT: NVIDIA® TensorRT™ is an SDK for high-performance deep learning inference on NVIDIA GPUs. This repository contains the open source components of TensorRT. and copied the repository link (GitHub - NVIDIA/TensorRT: NVIDIA® TensorRT™ is an SDK for high-performance deep learning inference on NVIDIA GPUs. This repository contains the open source components of TensorRT.) into the WebUI extension installation box, which installed successfully. However, upon restarting WebUI, the TensorRT tab is absent, and I encounter the following error upon launching WebUI:"

Python 3.10.11 (tags/v3.10.11:7d4cc5a, Apr 5 2023, 00:38:17) [MSC v.1929 64 bit (AMD64)]
Version: v1.9.4
Commit hash: feee37d75f1b168768014e4634dcb156ee649c05
Installing sd-webui-controlnet requirement: changing opencv-python version from 4.10.0.84 to 4.8.0
Launching Web UI with arguments: --xformers --api --skip-python-version-check --gradio-allowed-path C:\Stability Matrix\Data\Images
*** Error loading script: convert_te_onnx_to_trt_onnx.py
Traceback (most recent call last):
File “C:\Stability Matrix\Data\Packages\Stable Diffusion WebUI\modules\scripts.py”, line 508, in load_scripts
script_module = script_loading.load_module(scriptfile.path)
File “C:\Stability Matrix\Data\Packages\Stable Diffusion WebUI\modules\script_loading.py”, line 13, in load_module
module_spec.loader.exec_module(module)
File “”, line 883, in exec_module
File “”, line 241, in _call_with_frames_removed
File “C:\Stability Matrix\Data\Packages\Stable Diffusion WebUI\extensions\TensorRT\scripts\convert_te_onnx_to_trt_onnx.py”, line 23, in
import onnx_graphsurgeon as gs
ModuleNotFoundError: No module named ‘onnx_graphsurgeon’

TensorRT Version: 8.5.3.1

GPU Type: Geforce RTX 3060 RTX

Nvidia Driver Version: 556.12

CUDA Version: Built on Tue_Feb_27_16:28:36_Pacific_Standard_Time_2024
Cuda compilation tools, release 12.4, V12.4.99
Build cuda_12.4.r12.4/compiler.33961263_0

CUDNN Version: 9.2.1

Operating System + Version: Windows 11 Pro

Python Version (if applicable): 3.10

Git: 2.4.5

I managed to make TensorRT work in my last Windows11 installation (after a week of trial and error). But now that I installed Win11 again, I stand before the same problem and I have no idea what I did last time to make it work…:(

Thanks for an suggestion.

2

Hi @dezorianguy ,
You should be able to pip install it along with the other dependencies: python3 -m pip install onnx_graphsurgeon onnx onnxruntime
If that doesn’t work, you can specify the index with --index-url https://pypi.ngc.nvidia.com
Thanks

1 Like