deepspeedai/DeepSpeed

Pull Requests Commits

Use new dlpack api; Formatting fixes (#7101)

tjruwase committed 281 days ago

a7ab4ab7

Remove workflows for very old torch versions (#7090)

loadams committed 281 days ago

8254e1b3

Update parallelism for nv-torch-latest/nightly tests due to more GPUs/runner (#7086)

loadams committed 281 days ago

f9632484

Update README.md with ICS '23 MoE paper link (#7087)

siddharth9820 committed 281 days ago

9da1b336

Pin transformers version on tests that use latest. (#7085)

loadams committed 281 days ago

5ebe9bdc

Improve inference tutorial docs (#7083)

loadams committed 281 days ago

9c4c4447

Add pyproject.toml into comments

loadams committed 281 days ago

f325c92a

Add DeepseekV3 AutoTP. (#7045)

Yejing-Lai committed 281 days ago

5ab85893

Revert "Handle special case of libuv for Windows (#7064)" (#7076)

loadams committed 281 days ago

432e71dc

Add more requirements

loadams committed 281 days ago

00addaa7

Update workflow to not do build isolation

loadams committed 281 days ago

355e9c5e

Replace calls to `python setup.py sdist` with `python -m build --sdist` (#7069)

loadams committed 281 days ago

80ff4201

Fix meta load tensor imcompatible issue (#7073)

Yejing-Lai committed 281 days ago

c91d2256

config torch to avoid graph breaks caused by logger (#6999)

ShellyNR committed 281 days ago

fd5273f3

Fix TOCTOU issues, switch to fstat (#7067)

loadams committed 281 days ago

ac52baec

Add all imports from setup.py to pyproject.toml

loadams committed 281 days ago

5a2eceae

Bug Fix for offload_states API (#7050)

U-rara committed 281 days ago

630864d3

Update README with info on newest accelerator (#7065)

loadams committed 281 days ago

aaf86d68

Handle special case of libuv for Windows (#7064)

loadams committed 281 days ago

aa63ba4f

Tecorigin sdaa accelerator (#6903)

siqi654321 committed 281 days ago

cc79d173

fix an outdated doc wrt CUDA_VISIBLE_DEVICES (#7058)

stas00 committed 281 days ago

c83ade6f

Update version.txt after 0.16.4 release (#7063)

loadams committed 281 days ago

41718ada

Fix, bf16 optimizer remove dup loop (#7054)

wukong1992 committed 281 days ago

01dd3187

add autoTP training zero2 tests (#7049)

inkcherry committed 281 days ago

c0f4235d

Rename aio_thread_count to intra_op_parallelism (#7056)

tjruwase committed 281 days ago

5b2f7135

Update to actually use legacy backend

loadams committed 281 days ago

6d837ca1

Need legacy backend to execute setup.py, though it executes it differently, so we aren't seeing cupy installed.

loadams committed 281 days ago

4e884634

Test with non legacy backend

loadams committed 281 days ago

2873a11f

nv-ds-chat breaks with latest transformers (#7052)

loadams committed 281 days ago

c22be1ab

Update setup.py handling of ROCm cupy (#7051)

loadams committed 281 days ago

5326873f

Newer Older