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=532819f386be489bee79e60e2b9878f8fc403e5b
05/03/2024 4:13 AM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ARG
NVIDIA_BUILD_REF
05/03/2024 4:13 AM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4LABEL
com.nvidia.build.id=91431255
05/03/2024 4:13 AM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ENV
NVIDIA_BUILD_ID=91431255
05/03/2024 4:13 AM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ARG
NVIDIA_BUILD_ID
05/03/2024 4:13 AM UTC
sha256:a1bb87b1bdf02f2db521b87d05f1889788286c3de9fc4b7212f32b632f8cc877COPY
entrypoint.d/ /opt/nvidia/entrypoint.d/
05/03/2024 4:13 AM UTC
sha256:28e3223edf8b91e9d9553987db0199a226948cf8c6e2c3d4070359594012368eRUN
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 ${_CUDA_COMPAT_PATH}/lib
05/03/2024 4:13 AM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ENV
CUDA_MODULE_LOADING=LAZY
05/03/2024 4:13 AM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ENV
TORCH_CUDNN_V8_API_ENABLED=1
05/03/2024 4:13 AM UTC
sha256:8cd7a829f9590817f5a94f1e50408777dba173caa39f5471a1c4e2d8f5b01696RUN
PYVER=3.10 if [ "${L4T}" = "1" ]; then echo "Not installing Transformer Engine in iGPU container until Version variable is set"; else pip install --no-cache-dir --no-build-isolation git+https://github.com/NVIDIA/TransformerEngine.git@release_v${TRANSFORMER_ENGINE_VERSION}; fi
05/03/2024 4:13 AM UTC
...