mirror of
https://github.com/NVIDIA/TensorRT-LLM.git
synced 2026-01-13 22:18:36 +08:00
79 lines
1.8 KiB
Plaintext
79 lines
1.8 KiB
Plaintext
--extra-index-url https://download.pytorch.org/whl/cu130
|
|
-c constraints.txt
|
|
accelerate>=1.7.0
|
|
build
|
|
colored
|
|
cuda-python>=13
|
|
diffusers>=0.27.0
|
|
lark
|
|
mpi4py
|
|
numpy<2
|
|
onnx>=1.18.0,<1.20.0
|
|
onnx_graphsurgeon>=0.5.2
|
|
openai
|
|
polygraphy
|
|
psutil
|
|
nvidia-ml-py>=13
|
|
pulp
|
|
pandas
|
|
h5py==3.12.1
|
|
StrEnum
|
|
sentencepiece>=0.1.99
|
|
tensorrt~=10.13.3
|
|
# https://docs.nvidia.com/deeplearning/frameworks/pytorch-release-notes/rel-25-10.html#rel-25-10 uses 2.9.0a0.
|
|
torch>=2.9.0a0,<=2.9.0
|
|
torchvision
|
|
nvidia-modelopt[torch]~=0.37.0
|
|
# https://docs.nvidia.com/deeplearning/frameworks/pytorch-release-notes/rel-25-10.html#rel-25-10 uses 2.27.7
|
|
nvidia-nccl-cu13==2.27.7
|
|
nvidia-cuda-nvrtc
|
|
transformers==4.56.0
|
|
prometheus_client
|
|
prometheus_fastapi_instrumentator
|
|
pydantic>=2.9.1
|
|
pydantic-settings[yaml]
|
|
omegaconf
|
|
pillow
|
|
wheel<=0.45.1
|
|
optimum
|
|
# evaluate needs datasets>=2.0.0 which triggers datasets>3.1.0 which is not stable: https://github.com/huggingface/datasets/issues/7467
|
|
datasets==3.1.0
|
|
evaluate
|
|
mpmath>=1.3.0
|
|
click
|
|
click_option_group
|
|
aenum
|
|
pyzmq
|
|
fastapi>=0.120.1,<=0.121.3
|
|
starlette>=0.49.1
|
|
uvicorn
|
|
setuptools<80
|
|
ordered-set
|
|
peft
|
|
patchelf
|
|
einops
|
|
flashinfer-python>=0.3.0,<0.4.0
|
|
opencv-python-headless
|
|
xgrammar==0.1.25
|
|
llguidance==0.7.29
|
|
jsonschema
|
|
backoff
|
|
nvtx
|
|
matplotlib # FIXME: this is added to make nvtx happy
|
|
meson
|
|
ninja
|
|
etcd3 @ git+https://github.com/kragniz/python-etcd3.git@e58a899579ba416449c4e225b61f039457c8072a
|
|
blake3
|
|
soundfile
|
|
triton==3.5.0
|
|
tiktoken
|
|
blobfile
|
|
openai-harmony==0.0.4
|
|
nvidia-cutlass-dsl==4.3.1; python_version >= "3.10"
|
|
plotly
|
|
numexpr<2.14.0 # WAR for attempted use of nonexistent numpy.typing
|
|
partial_json_parser
|
|
apache-tvm-ffi==0.1.4 # used for reduce nvidia-cutlass-dsl host overhead
|
|
torch-c-dlpack-ext==0.1.3 # used for reduce nvidia-cutlass-dsl host overhead, optional package for improved torch tensor calling perf
|
|
mistral-common==1.8.6
|