Skip to main content
NVIDIA
TensorRT LLM Develop
Container
NVIDIA
TensorRT LLM Develop

TensorRT LLM provides users with an easy-to-use Python API to define Large Language Models (LLMs) and supports state-of-the-art optimizations to perform inference efficiently on NVIDIA GPUs.

  • LayerLabelCreated
    sha256:0462a37baeb6dfd4bf389e58f3f059f690c79da3cf6e3871273c450ad4e42c02RUN
    SH_ENV=/etc/shinit_v2 BASH_ENV=/etc/bash.bashrc GITHUB_MIRROR=https://urm.nvidia.com/artifactory/github-go-remote PYTHON_VERSION=3.12.3 CUDA_VER= CUDNN_VER= NCCL_VER= CUBLAS_VER= TORCH_INSTALL_TYPE=skip GIT_COMMIT=d210366ce69aa8d0dfc798395af917906858a228 TRT_LLM_VER=1.3.0rc25.dev2026082800 TARGETARCH=amd64 /bin/bash -c bash /tmp/gen_attribution.sh "devel" "${TRT_LLM_VER}" "${TARGETARCH}"
    08/28/2026 12:21 AM UTC
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ENV
    TRT_LLM_GIT_COMMIT=d210366ce69aa8d0dfc798395af917906858a228 TRT_LLM_VERSION=1.3.0rc25.dev2026082800
    08/28/2026 12:21 AM UTC
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ARG
    TARGETARCH=amd64
    08/28/2026 12:21 AM UTC
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ARG
    TRT_LLM_VER=1.3.0rc25.dev2026082800
    08/28/2026 12:21 AM UTC
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ARG
    GIT_COMMIT=d210366ce69aa8d0dfc798395af917906858a228
    08/28/2026 12:21 AM UTC
    sha256:92eef5e4d9fae272d4a52777c91c9bb9deb32ab20124709a8d2b26d6c1a0c972RUN
    SH_ENV=/etc/shinit_v2 BASH_ENV=/etc/bash.bashrc GITHUB_MIRROR=https://urm.nvidia.com/artifactory/github-go-remote PYTHON_VERSION=3.12.3 CUDA_VER= CUDNN_VER= NCCL_VER= CUBLAS_VER= TORCH_INSTALL_TYPE=skip /bin/bash -c GITHUB_MIRROR=${GITHUB_MIRROR} bash /opt/docker/common/install_ucx.sh &&
      GITHUB_MIRROR=${GITHUB_MIRROR} bash /opt/docker/common/install_nixl.sh &&
      bash /opt/docker/common/install_etcd.sh &&
      GITHUB_MIRROR=${GITHUB_MIRROR} bash /opt/docker/common/install_fa4.sh &&
      rm -rf /root/.cache/uv/archive-v0 &&
      rm -rf /usr/local/lib/python3.12/dist-packages/setuptools/_vendor/jaraco.context-5.3.0.dist-info &&
      rm -rf /usr/local/lib/python3.12/dist-packages/setuptools/_vendor/wheel-*.dist-info
    08/28/2026 12:21 AM UTC
    sha256:930e91e705df1c0fbf6c966b3388a621be0f5902bd74e8aadc3012e066713808RUN
    SH_ENV=/etc/shinit_v2 BASH_ENV=/etc/bash.bashrc GITHUB_MIRROR=https://urm.nvidia.com/artifactory/github-go-remote PYTHON_VERSION=3.12.3 CUDA_VER= CUDNN_VER= NCCL_VER= CUBLAS_VER= TORCH_INSTALL_TYPE=skip /bin/bash -c pip3 uninstall -y tornado black nbconvert pillow nvidia-cutlass-dsl nvidia-cutlass-dsl-libs-base nvidia-cutlass-dsl-libs-core nvidia-cutlass-dsl-libs-cu12 nvidia-cutlass-dsl-libs-cu13 numpy wandb ||
      true &&
      rm -rf $(python3 -c "import site; print(site.getsitepackages()[0])")/nvidia_cutlass_dsl* &&
      pip3 install -r /tmp/constraints.txt "nvidia-cutlass-dsl[cu13]==4.6.1" &&
      rm /tmp/constraints.txt
    08/28/2026 12:13 AM UTC
    sha256:22359ca77f4405c009f057d2c94dd8bdc64a0e4a718db33a860c0aa9d4468216COPY
    constraints.txt /tmp/constraints.txt
    08/28/2026 12:12 AM UTC
    sha256:2bfb0bea8e0fe817b8f363ddf48f9197ead1e1538106b5a962a960cd4a9fa721RUN
    SH_ENV=/etc/shinit_v2 BASH_ENV=/etc/bash.bashrc GITHUB_MIRROR=https://urm.nvidia.com/artifactory/github-go-remote PYTHON_VERSION=3.12.3 CUDA_VER= CUDNN_VER= NCCL_VER= CUBLAS_VER= TORCH_INSTALL_TYPE=skip /bin/bash -c echo "Using GitHub mirror: ${GITHUB_MIRROR}" &&
      echo "Using Python version: ${PYTHON_VERSION}" &&
      GITHUB_MIRROR=${GITHUB_MIRROR} PYTHON_VERSION=${PYTHON_VERSION} CUDA_VER=${CUDA_VER} CUDNN_VER=${CUDNN_VER} NCCL_VER=${NCCL_VER} CUBLAS_VER=${CUBLAS_VER} TORCH_INSTALL_TYPE=${TORCH_INSTALL_TYPE} bash /opt/docker/common/install.sh --base --cmake --ccache --cuda_toolkit --cuda_libs --polygraphy --mpi4py --pytorch
    08/28/2026 12:12 AM UTC
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ARG
    TORCH_INSTALL_TYPE=skip
    08/28/2026 12:08 AM UTC
    ...

    NVIDIA uses cookies to improve your experience on our web site. We and our third-party partners also use cookies and other tools to collect and record information you provide as well as information about your interactions with our websites for performance improvement, analytics, and to assist in marketing efforts. By clicking "Accept All", you consent to our use of cookies and other tools as described in our Cookie Policy. You can manage your cookie settings by clicking on "Manage Settings." By continuing to use this site or by clicking one of the buttons below, you agree to our Terms of Service (which contains important waivers). Please see our Privacy Policy for more information on our privacy practices.