代码拉取完成,页面将自动刷新
--extra-index-url https://pypi.nvidia.com
accelerate==0.25.0
build
colored
cuda-python==12.3.0
diffusers==0.15.0
mpi4py
numpy
onnx>=1.12.0
polygraphy
psutil
pynvml>=11.5.0
pulp
pandas
h5py
StrEnum
pywin32
StrEnum
sentencepiece>=0.1.99
# WAR the new posting of "nvidia-cudnn-cu12~=9.0".
# "tensorrt==9.3.0.post12.dev1" specifies "nvidia-cudnn-cu12" but actually requires "nvidia-cudnn-cu12~=8.9".
nvidia-cudnn-cu12~=8.9; platform_machine == "x86_64"
tensorrt==9.3.0.post12.dev1
tokenizers>=0.14
# Default torch is CPU-only on Windows, so need to specify a torch version with GPU support
torch==2.2.0+cu121
transformers==4.38.2
wheel
optimum
evaluate
janus
mpmath==1.3.0
此处可能存在不合适展示的内容,页面不予展示。您可通过相关编辑功能自查并修改。
如您确认内容无涉及 不当用语 / 纯广告导流 / 暴力 / 低俗色情 / 侵权 / 盗版 / 虚假 / 无价值内容或违法国家有关法律法规的内容,可点击提交进行申诉,我们将尽快为您处理。