Segmentation fault when following the Phi-3 tutorial for DirectML

[matt200-ok]

Describe the issue

When following the Phi-3 Tutorial using DirectML the phi3-qa.py script causes a segmentation fault after entering the input.

This issue occurs a laptop with a 13th Gen Intel(R) Core(TM) i7-1360P and a laptop with an AMD Ryzen™ AI 9 HX 370.

To reproduce

• Follow the Phi-3 Tutorials Run With DirectML section.
  • huggingface-cli download microsoft/Phi-3-mini-4k-instruct-onnx --include directml/* --local-dir .
  • pip install numpy
  • pip install --pre onnxruntime-genai-directml
  • curl https://raw.githubusercontent.com/microsoft/onnxruntime-genai/main/examples/python/phi3-qa.py -o phi3-qa.py
  • python phi3-qa.py -m directml\directml-int4-awq-block-128
    • Ask question such as Tell me a joke about GPUs
• Segmentation Fault.

Urgency

No response

Platform

Windows

OS Version

Windows 11 Home 22631.4460

ONNX Runtime Installation

Released Package

ONNX Runtime Version or Commit ID

1.20.0

ONNX Runtime API

Python

Architecture

X64

Execution Provider

DirectML

Execution Provider Library Version

0.5.1