amd-quark-llama-tiny-w-int8-b-int8-per-tensor_ref_output.pt
fxmarty-llama-tiny-w-int8-b-int8-per-tensor_ref_output.pt