Cannot create TensorRT session in Java without using addCUDA first #26493
Description
Describe the issue
I'm attempting to load a model to the TensorRTExecutionProvider using the Java API but I am getting an error unless I use addCUDA prior to addTensorrt. Though this avoids the error, the model is loaded to the CUDAExecutionProvider instead of TensorRTExecutionProvider
The error I get is below
2025-11-04 11:10:47.790429743 [E:onnxruntime:ort-java, provider_bridge_ort.cc:2195 TryGetProviderInfo_CUDA] /onnxruntime_src/onnxruntime/core/session/provider_bridge_ort.cc:1778 onnxruntime::Provider& onnxruntime::ProviderLibrary::Get() [ONNXRuntimeError] : 1 : FAIL : Failed to load library libonnxruntime_providers_cuda.so with error: libonnxruntime_providers_cuda.so: cannot open shared object file: No such file or directory
Exception in thread "main" ai.onnxruntime.OrtException: Error code - ORT_FAIL - message: /onnxruntime_src/onnxruntime/core/session/provider_bridge_ort.cc:2203 onnxruntime::ProviderInfo_CUDA& onnxruntime::GetProviderInfo_CUDA() CUDA Provider not available, can't get interface for it
at ai.onnxruntime.providers.OrtTensorRTProviderOptions.applyToNative(Native Method)
at ai.onnxruntime.providers.StringConfigProviderOptions.applyToNative(StringConfigProviderOptions.java:90)
at ai.onnxruntime.OrtSession$SessionOptions.addTensorrt(OrtSession.java:1182)
...
To reproduce
Attempt to load a model using SessionOptions.addTensorrt
Urgency
No response
Platform
Linux
OS Version
Ubuntu 22.04.5 LTS
ONNX Runtime Installation
Released Package
ONNX Runtime Version or Commit ID
1.22.0
ONNX Runtime API
Java
Architecture
X86
Execution Provider
TensorRT
Execution Provider Library Version
TensorRT 10.9