Hi, I try to run the sample of this.
I believe that the "dml" in provider_options would use the GPU to compute.
However, I get this error while entry the prompt:
Microsoft.ML.OnnxRuntimeGenAI.OnnxRuntimeGenAIException: 887A0020 An internal issue prevented the driver from carrying out the specified operation. The driver's state is probably suspect
By the way, I have testified that without the "dml" in provider_options, everything work fine and super slow.
Other question:
How do I choose to run the computation on my NPU?
Environment:
Windows 11 Pro 24H2
CPU: Snapdragon(R) X Elite - X1E78100 - Qualcomm(R) Oryon(TM) CPU 3.42 GHz
GPU: Snapdragon(R) X Elite - X1E78100 - Qualcomm(R) Adreno(TM) GPU
NPU: Snapdragon(R) X Elite - X1E78100 - Qualcomm(R) Hexagon(TM) NPU
I am wondering that the model of phi-3 is currently not support for Qualcomm GPU and NPU
Thanks for any help! :D
Hi, I try to run the sample of this. I believe that the "dml" in provider_options would use the GPU to compute. However, I get this error while entry the prompt:
Microsoft.ML.OnnxRuntimeGenAI.OnnxRuntimeGenAIException: 887A0020 An internal issue prevented the driver from carrying out the specified operation. The driver's state is probably suspect
By the way, I have testified that without the "dml" in provider_options, everything work fine and super slow.
Other question: How do I choose to run the computation on my NPU?
Environment: Windows 11 Pro 24H2 CPU: Snapdragon(R) X Elite - X1E78100 - Qualcomm(R) Oryon(TM) CPU 3.42 GHz GPU: Snapdragon(R) X Elite - X1E78100 - Qualcomm(R) Adreno(TM) GPU NPU: Snapdragon(R) X Elite - X1E78100 - Qualcomm(R) Hexagon(TM) NPU
I am wondering that the model of phi-3 is currently not support for Qualcomm GPU and NPU Thanks for any help! :D