Nicolas Mowen 81d7c47129 Optimize OpenVINO and ONNX Model Runners (#20063) ...

* Use re-usable inference request to reduce CPU usage

* Share tensor

* Don't count performance

* Create openvino runner class

* Break apart onnx runner

* Add specific note about inability to use CUDA graphs for some models

* Adjust rknn to use RKNNRunner

* Use optimized runner

* Add support for non-complex models for CudaExecutionProvider

* Use core mask for rknn

* Correctly handle cuda input

* Cleanup

* Sort imports

2025-09-14 06:22:22 -06:00

..

onnx

Optimize OpenVINO and ONNX Model Runners (#20063)

2025-09-14 06:22:22 -06:00

__init__.py

…

embeddings.py

Enrichments: Allow targeting a specific GPU ID (#19342)

2025-08-18 17:43:53 -06:00

maintainer.py

…

util.py

…