CPU used instead of GPU

[7lano]

I am using an NVIDIA RTX 3080 Ti, but the application always falls back to CPU instead of using the GPU.

Error log (anonymized):
2025-08-22 13:46:58.7563265 [E:onnxruntime:Default, provider_bridge_ort.cc:2195 onnxruntime::TryGetProviderInfo_CUDA] D:\a\_work\1\s\onnxruntime\core\session\provider_bridge_ort.cc:1778 onnxruntime::ProviderLibrary::Get [ONNXRuntimeError] : 1 : FAIL : Error loading "...\AppData\Local\Programs\Python\Python313\Lib\site-packages\onnxruntime\capi\onnxruntime_providers_cuda.dll" which depends on "cudnn64_9.dll" which is missing.

System details:

GPU: NVIDIA RTX 3080 Ti
OS: Windows 11
Python: 3.13.7
Rembg-Fuse: commit from Aug 21, 2025

[Akascape]

@7lano
Run these commands in the terminal window, and let me know:

• Uninstall onnxruntime using
  pip uninstall onnxruntime

• Install onnxruntime-gpu using
  pip install onnxruntime-gpu

[dave9123]

Same issue for me apparently

Resolve_VgdJK6ZbnA.mp4

NVIDIA driver version: 581.7

> pip freeze
aiofiles==24.1.0
aiohappyeyeballs==2.6.1
aiohttp==3.13.2
aiosignal==1.4.0
annotated-doc==0.0.4
annotated-types==0.7.0
anyio==4.11.0
asyncer==0.0.10
attrs==25.4.0
audioop-lts==0.2.2
brotli==1.2.0
certifi==2025.11.12
charset-normalizer==3.4.4
click==8.3.1
colorama==0.4.6
coloredlogs==15.0.1
fastapi==0.122.0
ffmpy==1.0.0
filelock==3.20.0
filetype==1.2.0
flatbuffers==25.9.23
frozenlist==1.8.0
fsspec==2025.10.0
gradio==6.0.1
gradio_client==2.0.0
groovy==0.1.2
h11==0.16.0
hf-xet==1.2.0
httpcore==1.0.9
httpx==0.28.1
huggingface_hub==1.1.5
humanfriendly==10.0
idna==3.11
ImageIO==2.37.2
Jinja2==3.1.6
jsonschema==4.25.1
jsonschema-specifications==2025.9.1
lazy_loader==0.4
llvmlite==0.45.1
markdown-it-py==4.0.0
MarkupSafe==3.0.3
mdurl==0.1.2
more-itertools==10.8.0
mpmath==1.3.0
multidict==6.7.0
networkx==3.6
numba==0.62.1
numpy==2.2.6
onnxruntime-gpu==1.23.2
openai-whisper==20250625
opencv-python-headless==4.12.0.88
orjson==3.11.4
packaging==25.0
pandas==2.3.3
pillow==12.0.0
platformdirs==4.5.0
pooch==1.8.2
propcache==0.4.1
protobuf==6.33.1
pydantic==2.12.4
pydantic_core==2.41.5
pydub==0.25.1
Pygments==2.19.2
PyMatting==1.1.14
pyreadline3==3.5.4
python-dateutil==2.9.0.post0
python-multipart==0.0.20
pytz==2025.2
PyYAML==6.0.3
referencing==0.37.0
regex==2025.11.3
rembg==2.0.68
requests==2.32.5
rich==14.2.0
rpds-py==0.29.0
safehttpx==0.1.7
scikit-image==0.25.2
scipy==1.16.3
semantic-version==2.10.0
setuptools==80.9.0
shellingham==1.5.4
six==1.17.0
sniffio==1.3.1
starlette==0.50.0
sympy==1.14.0
tifffile==2025.10.16
tiktoken==0.12.0
tomlkit==0.13.3
torch==2.9.1
tqdm==4.67.1
typer==0.20.0
typer-slim==0.20.0
typing-inspection==0.4.2
typing_extensions==4.15.0
tzdata==2025.2
urllib3==2.5.0
uv==0.9.13
uvicorn==0.38.0
watchdog==6.0.0
yarl==1.22.0

[senavi888]

@7lano Run these commands in the terminal window, and let me know:

* Uninstall onnxruntime using
  `pip uninstall onnxruntime`

* Install onnxruntime-gpu using
  `pip install onnxruntime-gpu`

I have the same problem. I uninstalled onnxruntime, but it wasn't even installed. I tried installing onnxruntime-gpu, and it already installed.

[senavi888]

I'm finded problem. Need to use specific versions of the surrounding software, not the latest ones, and check what's set in the environment variables. The error disappeared for me, and everything started being processed by the GPU after that.

Python 3.10.11 
Cuda toolkit 12.9
cudnn 9.16 

And most impotant you need manualy add Environment variables (SYSTEM) > Path:

C:\Program Files\NVIDIA\CUDNN\v9.16\bin\12.9
C:\Program Files\NVIDIA\CUDNN\v9.16\lib\12.9\x64