microsoft/DeepSpeed

Pull Requests Commits

Merge branch 'master' into loadams/cuda-compilation-nv-bfloat162

loadams committed 35 days ago

Verified 134399e2

Update CPU torch version to 2.7 (#7241)

loadams committed 35 days ago

Verified 9926879b

Revert "Update torch cpu test version"

loadams committed 36 days ago

8d2865e0

Update torch cpu test version

loadams committed 36 days ago

00b5678b

Merge branch 'master' into loadams/cuda-compilation-nv-bfloat162

loadams committed 38 days ago

Verified f2b560b5

Add cpu accelerator fp16 dtype support (#7207)

Yejing-Lai committed 38 days ago

Verified d79bd930

[NFC] Fix comment related to SP group (#7234)

c8ef committed 38 days ago

Verified 9c9d32c2

Recommend using latest (#7233)

tohtana committed 41 days ago

Verified 962a8f0a

Update version.txt after 0.16.7 release (#7232)

loadams committed 41 days ago

Verified ff231af7

Make sure it's not None before offloading contiguous_grad_buffer (#7227)

HollowMan6 committed 41 days ago

Verified c66fdaf3

Pass `with_cuda` arg for jit_load in OpBuilder (#7226)

HollowMan6 committed 41 days ago

Verified 625ea528

Add defence for DeepCompile w/o optimizer (#7225)

HollowMan6 committed 42 days ago

Verified 86e51e69

Fix build on AMD GPUs (related to DeepCompile) (#7224)

HollowMan6 committed 42 days ago

Verified 0f224d79

Fix pass for z3 and profiler (#7222)

tohtana committed 42 days ago

Verified 1d022e05

Fix release links (#7219)

tjruwase committed 43 days ago

Verified 8f93f8b9

Update version.txt after 0.16.6 release (#7218)

loadams committed 43 days ago

Verified a01a2688

DeepCompile for enhanced compiler integration (#7154)

tohtana committed 43 days ago

Verified 227a60c0

Merge branch 'master' into loadams/cuda-compilation-nv-bfloat162

hwchen2017 committed 49 days ago

Verified 989fb619

Add defence for offload_states and reload_states w/o optimizer (#7211)

HollowMan6 committed 49 days ago

Verified a21e5b9d

Support complicated use cases with TiedLayerSpec (#7208)

limjcst committed 50 days ago

Verified 185330cd

HPU accelerator memory mapping is broken because of torch fill uninit memory (#7209)

oelayan7 committed 51 days ago

Verified 56005d2b

update dependencies version info (#7206)

inkcherry committed 51 days ago

Verified 29fa95a8

Merge branch 'master' into loadams/cuda-compilation-nv-bfloat162

loadams committed 55 days ago

Verified f66296dc

Update to fix pydantic warning (#7193)

loadams committed 56 days ago

Verified 027ee21a

Reland perf fix for nan inf check (#7184)

nelyahu committed 57 days ago

Verified 3c1817f3

Update to new torch grad hook API: BF16Optimizer and Stage2 (#7189)

deepcharm committed 57 days ago

Verified 79ff1627

Add qwen3 autotp support (#7187)

Yejing-Lai committed 57 days ago

Verified 11aca297

Remove definition that causes error in cuda 12.6

loadams committed 59 days ago

fbaacf35

Fix issue #5242 grad_norm and loss is nan (#7171)

Glaceon-Hyy committed 61 days ago

Verified 1f706621

async tp allreduce (#7115)

inkcherry committed 61 days ago

Verified b8cc1eb0

Older