PR #26002 Reduce Python and Nuget GPU package size

remove the oldest CUDA arch

chilo-ms committed 340 days ago

update

chilo-ms committed 339 days ago

revmoe older arch in Linux python whl, Linux nuget and windows nuget

chilo-ms committed 339 days ago

remove old GPU arch for windows nuget

chilo-ms committed 339 days ago

remove old GPU arch for python wheel

chilo-ms committed 339 days ago

reduce number of k support in beam search

chilo-ms committed 334 days ago

revert beam_search_topk

chilo-ms committed 333 days ago

remove FPA_INTB_GEMM for Linux python wheel

chilo-ms committed 333 days ago

revert beam_search_topk

chilo-ms committed 333 days ago

Add back support for SM75 on Linux and disable FPA_INTB_GEMM

chilo-ms committed 332 days ago

onnxruntime Reduce Python and Nuget GPU package size #26002 Merged