diff --git a/docs/benchmarks/language/llama2-70b.md b/docs/benchmarks/language/llama2-70b.md index e68693716..40c62cf71 100644 --- a/docs/benchmarks/language/llama2-70b.md +++ b/docs/benchmarks/language/llama2-70b.md @@ -25,3 +25,10 @@ hide: {{ mlperf_inference_implementation_readme (4, "llama2-70b-99", "neuralmagic") }} {{ mlperf_inference_implementation_readme (4, "llama2-70b-99.9", "neuralmagic") }} + +=== "AMD" + ## AMD MLPerf Implementation + +{{ mlperf_inference_implementation_readme (4, "llama2-70b-99", "amd") }} + +{{ mlperf_inference_implementation_readme (4, "llama2-70b-99.9", "amd") }} diff --git a/main.py b/main.py index 38e6e1650..8b6ce3ea5 100755 --- a/main.py +++ b/main.py @@ -43,6 +43,11 @@ def mlperf_inference_implementation_readme(spaces, model, implementation, *, imp devices = [ "CUDA" ] frameworks = [ "TensorRT" ] + elif implementation == "amd": + devices = [ "cuda" ] + frameworks = [ "pytorch" ] + execution_envs.remove("Docker") + elif implementation == "neuralmagic": devices = [ "CUDA" ] frameworks = [ "pytorch" ]