Skip to content

Pull requests: microsoft/DeepSpeed

Author
Filter by author
Label
Filter by label
Use alt + click/return to exclude labels
or + click/return for logical OR
Projects
Filter by project
Milestones
Filter by milestone
Reviews
Assignee
Filter by who’s assigned
Sort

Pull requests list

[NPU]Add ZeRO-Infinity feature for NPU
#4809 opened Dec 13, 2023 by misstek Loading…
Inference V2 Human Eval
#4804 opened Dec 12, 2023 by lekurile Draft
Update to latest torch
#4798 opened Dec 11, 2023 by loadams Loading…
[CI] Use latest transformers
#4796 opened Dec 11, 2023 by loadams Loading…
Unit tests for MiCS
#4792 opened Dec 9, 2023 by zarzen Loading…
fix falcon model load from_config meta_data error
#4783 opened Dec 7, 2023 by baodii Loading…
Test runsc docker on a6000
#4778 opened Dec 6, 2023 by loadams Loading…
Loading FalconLinear to meta
#4773 opened Dec 5, 2023 by oelayan7 Loading…
Support FP16 CpuAdam + Zero Stage 3
#4771 opened Dec 4, 2023 by lz1oceani Loading…
[DO NOT MERGE] Fix for accelerate unit tests
#4769 opened Dec 4, 2023 by mrwyattii Loading…
Universal Checkpoint for Sequence Parallelism
#4752 opened Nov 29, 2023 by samadejacobs Loading…
Update README.md Windows Instructions
#4748 opened Nov 28, 2023 by erew123 Loading…
Update flops profiler to handle attn and __matmul__
#4724 opened Nov 24, 2023 by KimmiShi Loading…
params partition for skip_init
#4722 opened Nov 23, 2023 by inkcherry Loading…
support baichuan model:
#4721 opened Nov 23, 2023 by baodii Loading…
fix confusing width in simd_load
#4714 opened Nov 22, 2023 by yzhblind Loading…
ProTip! no:milestone will show everything without a milestone.