Remove the tensorcores llama.cpp wheels

They are not faster than the default wheels anymore and they use a lot of space.
This commit is contained in:
oobabooga 2024-07-22 11:52:40 -07:00
parent 017d2332ea
commit a687f950ba
7 changed files with 1 additions and 19 deletions

View file

@ -46,7 +46,6 @@ loaders_and_params = OrderedDict({
'numa',
'no_offload_kqv',
'row_split',
'tensorcores',
'flash_attn',
'streaming_llm',
'attention_sink_size',
@ -73,7 +72,6 @@ loaders_and_params = OrderedDict({
'logits_all',
'no_offload_kqv',
'row_split',
'tensorcores',
'flash_attn',
'streaming_llm',
'attention_sink_size',