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
    PyTorch
    Container
    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.

  • LayerLabelCreated
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4LABEL
    com.nvidia.build.ref=b3a219ee36b8fb6bf33353e548d995ed552ccbfa
    		10/17/2025 2:16 AM UTC
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ARG
    NVIDIA_BUILD_REF=b3a219ee36b8fb6bf33353e548d995ed552ccbfa
    		10/17/2025 2:16 AM UTC
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4LABEL
    com.nvidia.build.id=222058776
    		10/17/2025 2:16 AM UTC
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ENV
    NVIDIA_BUILD_ID=222058776
    		10/17/2025 2:16 AM UTC
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ARG
    NVIDIA_BUILD_ID=222058776
    		10/17/2025 2:16 AM UTC
    sha256:5776ff2dcf33a85dd5dc06b2d1961ea0cd7d689be07601519b2cfd0cb20879f1RUN
    NVIDIA_PYTORCH_VERSION=25.10 PYTORCH_BUILD_VERSION=2.9.0a0+145a3a7 NVFUSER_BUILD_VERSION=ed11f2b TARGETARCH=arm64 PYVER=3.12 L4T=1 ENABLE_MITMPROXY=0 DALI_EXTRA_INDEX_URL=http://sqrl/nvdl/datasets/dali/pip-dali cd /opt/pytorch/pytorch/ &&
  sed -i "s#    if (props.computeMode != cudaComputeModeDefault) {#    int computeMode = -1;\n    TP_CUDA_CHECK(cudaDeviceGetAttribute(\&computeMode, cudaDevAttrComputeMode, device.index));\n    if (computeMode != cudaComputeModeDefault) {#g" ./third_party/tensorpipe/tensorpipe/channel/cuda_ipc/context_impl.cc &&
  cd -
    		10/17/2025 2:16 AM UTC
    sha256:012b8d075af47c5a5ab1d728ba53ea7cca5c366ae351fdaee22d92a8f587bd2dCOPY
    entrypoint.d/ /opt/nvidia/entrypoint.d/
    		10/17/2025 2:16 AM UTC
    sha256:576c77eaf319d88796b1cb8659047db0c2a051d9298b53ce4a363c594018fc76COPY
    constraint.txt /etc/pip/original_constraint.txt
    		10/17/2025 2:16 AM UTC
    sha256:697913a502810abf3e9756a9fd635891842b56017e048dc46dec532136e46a65COPY
    restricted_constraint.txt /etc/pip/constraint.txt
    		10/17/2025 2:16 AM UTC
    sha256:35cf0614ebeb2c0931c08941d7f56d7f5b165b505a44dda81c0d5aeea7878ff9RUN
    NVIDIA_PYTORCH_VERSION=25.10 PYTORCH_BUILD_VERSION=2.9.0a0+145a3a7 NVFUSER_BUILD_VERSION=ed11f2b TARGETARCH=arm64 PYVER=3.12 L4T=1 ENABLE_MITMPROXY=0 DALI_EXTRA_INDEX_URL=http://sqrl/nvdl/datasets/dali/pip-dali 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
    		10/17/2025 2:16 AM UTC
    ...