[Kernel] Use flash-attn for decoding (#3648)
Co-authored-by: Woosuk Kwon <woosuk.kwon@berkeley.edu> Co-authored-by: LiuXiaoxuanPKU <lilyliupku@gmail.com>
This commit is contained in:
co-authored by
Woosuk Kwon
LiuXiaoxuanPKU
parent
ce532ff45c
commit
1356df53bd
@@ -12,7 +12,7 @@ MODELS = [
|
||||
# "Deci/DeciLM-7b", # Broken
|
||||
# "tiiuae/falcon-7b", # Broken
|
||||
"EleutherAI/gpt-j-6b",
|
||||
"mosaicml/mpt-7b",
|
||||
# "mosaicml/mpt-7b", # Broken
|
||||
# "Qwen/Qwen1.5-0.5B" # Broken,
|
||||
]
|
||||
|
||||
|
||||
Reference in New Issue
Block a user