ContainersNVIDIADeepSeek-R1NVIDIALlama-3.1-Nemotron-70B-InstructNVIDIAPyTorch
CollectionsNVIDIAOmniverse Kit (FB)NVIDIADeepStream SDKNVIDIAOmniverse Kit App Streaming
ModelsNVIDIAStyleGAN3 pretrained modelsNVIDIAPeopleNetNVIDIATrafficCamNet
ResourcesNVIDIARiva Skills Quick Start
Helm ChartsNVIDIAGPU OperatorNVIDIANVIDIA NIM Operator
NVIDIA
NVIDIA
PyTorch
Container
NVIDIA
NVIDIA
PyTorch

PyTorch is a GPU accelerated tensor computational framework. Functionality can be extended with common Python libraries such as NumPy and SciPy. Automatic differentiation is done with a tape-based system at the functional and neural network layer levels.

NVIDIA AI Enterprise Supported
LayerLabelCreated
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4LABEL
com.nvidia.build.ref=3e3c067dd015e6d16d2cf59ac18e9f2e2466b68a
10/03/2024 9:46 PM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ARG
NVIDIA_BUILD_REF=3e3c067dd015e6d16d2cf59ac18e9f2e2466b68a
10/03/2024 9:46 PM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4LABEL
com.nvidia.build.id=114410972
10/03/2024 9:46 PM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ENV
NVIDIA_BUILD_ID=114410972
10/03/2024 9:46 PM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ARG
NVIDIA_BUILD_ID=114410972
10/03/2024 9:46 PM UTC
sha256:2ee074064687eb16f49d37e2a3d049e25cec0dffccf3dc6d296275d25de09eb5COPY
entrypoint.d/ /opt/nvidia/entrypoint.d/
10/03/2024 9:46 PM UTC
sha256:508420e7ba944227fce0dc5b925a106561efbc608082a2d7f093ff6a45956a2eRUN
NVIDIA_PYTORCH_VERSION=24.10 PYTORCH_BUILD_VERSION=2.5.0a0+e000cf0 NVFUSER_BUILD_VERSION=f669fcf TARGETARCH=amd64 PYVER=3.10 L4T=0 PYVER=3.10 ln -sf ${_CUDA_COMPAT_PATH}/lib.real ${_CUDA_COMPAT_PATH}/lib &&
  echo ${_CUDA_COMPAT_PATH}/lib > /etc/ld.so.conf.d/00-cuda-compat.conf &&
  ldconfig &&
  rm -f ${_CUDA_COMPAT_PATH}/lib
10/03/2024 9:46 PM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ENV
CUDA_MODULE_LOADING=LAZY
10/03/2024 9:46 PM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ENV
TORCH_CUDNN_V8_API_ENABLED=1
10/03/2024 9:46 PM UTC
sha256:f5b617408786d6278148ff0792831215d2914e25718fe385e0ce206fbcd13f80RUN
NVIDIA_PYTORCH_VERSION=24.10 PYTORCH_BUILD_VERSION=2.5.0a0+e000cf0 NVFUSER_BUILD_VERSION=f669fcf TARGETARCH=amd64 PYVER=3.10 L4T=0 PYVER=3.10 if [ "${L4T}" = "1" ]; then echo "Not installing Transformer Engine in iGPU container until Version variable is set"; else NVTE_BUILD_THREADS_PER_JOB=8 pip install --no-cache-dir --no-build-isolation git+https://github.com/NVIDIA/TransformerEngine.git@release_v${TRANSFORMER_ENGINE_VERSION}; fi
10/03/2024 9:46 PM UTC
...