DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.
COMMITS
/ csrc/quantization/pt_binding.cpp December 10, 2024
X
Merge LoCo with Zero++ (#6730)
xyxie committed
May 31, 2024
C
Fixed the Windows build. (#5596)
Costin Eseanu committed
October 11, 2023
September 11, 2023
O
ZeRO-Inference refresh (#4197)
Olatunji Ruwase committed
August 9, 2023
C
Pass correct node size for ZeRO++ (#4085)
Connor Holmes committed
June 23, 2023
H
ZeRO++ (#3784)
Heyang Qin committed
March 31, 2023
M
Update DeepSpeed copyright license to Apache 2.0 (#3111)
Michael Wyatt committed
January 12, 2023
L
Extend quantization utils features (#2683)
LOK CHAND KOPPAKA committed
December 8, 2022
L
Support N-dimension input in quantization kernel (#2575)
lokoppakmsft committed
November 19, 2022
C
Initial dequant library implementation (#2521)
Connor Holmes committed
November 17, 2022
L
Deepspeed quantization library v0.1 (#2450)
lokoppakmsft committed
October 2, 2021
A
Fix many typos (#1423)
Alex Hedges committed
May 24, 2021
R
Quantization + inference release (#1091)
Reza Yazdani committed