openteamsinc
pypi/deepspeed
Risk Profile
Package
Source
Risk Profile
Maturity:
Mature
Health:
Healthy
Legal:
Moderate Risk
License includes a patent grant clause
Package license does not match the source code license
Security:
Caution Needed
At least one high or critical severity vulnerability has been reported in the last 600
Package
pypi
deepspeed
Version
0.18.4
Last Release Date
11 days ago
License
Apache Software License 2.0
Dependencies
einops
hjson
msgpack
ninja
numpy
packaging
>=20.0
psutil
py-cpuinfo
pydantic
>=2.0.0
torch
tqdm
1bit-mpi:
mpi4py
readthedocs:
autodoc_pydantic
>=2.0.0
docutils
<0.18
hjson
packaging
psutil
py-cpuinfo
pydantic
>=2.0.0
recommonmark
sphinx_rtd_theme
torch
tqdm
dev:
accelerate
clang-format
==18.1.3
comet_ml
>=3.41.0
deepspeed-kernels
docutils
<0.18
future
importlib-metadata
>=4
mup
pre-commit
>=3.2.0
pytest
<8.4.0, >=7.2.0
pytest-forked
pytest-randomly
pytest-xdist
qtorch
==0.3.0
recommonmark
sphinx
sphinx-rtd-theme
tensorboard
torchvision
transformers
>=4.51.3
wandb
autotuning:
tabulate
autotuning-ml:
hjson
tabulate
xgboost
sparse-attn:
triton
==1.0.0
sparse:
neural-compressor
==2.1.0
inf:
google
lm-eval
==0.3.0
protobuf
qtorch
safetensors
sentencepiece
transformers
>=4.32.1
sd:
diffusers
>=0.25.0
triton
>=2.1.0
triton:
triton
==2.1.0
deepcompile:
scipy
all:
pydantic
>=2.0.0
protobuf
transformers
>=4.32.1
clang-format
==18.1.3
pytest-forked
lm-eval
==0.3.0
psutil
autodoc_pydantic
>=2.0.0
future
sphinx-rtd-theme
neural-compressor
==2.1.0
mpi4py
tensorboard
sphinx_rtd_theme
torchvision
transformers
>=4.51.3
scipy
tabulate
deepspeed-kernels
accelerate
sphinx
wandb
qtorch
diffusers
>=0.25.0
importlib-metadata
>=4
packaging
safetensors
comet_ml
>=3.41.0
mup
pytest-xdist
google
xgboost
pytest
<8.4.0, >=7.2.0
recommonmark
tqdm
triton
==2.1.0
torch
sentencepiece
hjson
pre-commit
>=3.2.0
docutils
<0.18
triton
==1.0.0
qtorch
==0.3.0
pytest-randomly
py-cpuinfo
triton
>=2.1.0
Source
Location
deepspeedai/DeepSpeed
Last Source Update
2 days ago
Licenses
Apache License 2.0
Apache License 2.0
Distribution Destinations
pypi/deepspeed