AMD-SHARK-Studio/amdshark/examples/amdshark_inference/llama

AMDSHARK LLaMA

TORCH-MLIR Version

https://github.com/nod-ai/torch-mlir.git

Then check out the complex branch and git submodule update --init and then build with .\build_tools\python_deploy\build_windows.ps1

Setup & Run

git clone https://github.com/nod-ai/llama.git

Then in this repository

pip install -e .
python llama/amdshark_model.py