NVIDIA
NVIDIA
Dynamo vLLM Runtime
Container
NVIDIA
NVIDIA
Dynamo vLLM Runtime

The Dynamo vLLM runtime image is a containerized build of Dynamo + vLLM which serves as the base runtime environment for vLLM based inference with Dynamo's distributed inference framework.

  • LayerLabelCreated
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4USER
    dynamo
    06/12/2026 8:26 PM UTC
    sha256:22ad408e38ee2e208b2f264e067bf461c74ae2c29d421df8b43db4732f24be85RUN
    MODELEXPRESS_VERSION=0.4.0 DEVICE=cuda PYTHON_VERSION=3.12 SITE_PACKAGES=/opt/dynamo/venv/lib/python3.12/site-packages ENABLE_KVBM=true ENABLE_GPU_MEMORY_SERVICE=true DYNAMO_COMMIT_SHA=919682da679aa699d5bca9c872f4c1d9a530bbc0 EFA_VERSION=1.47.0 NIXL_LIBFABRIC_REF=v2.5.1 /bin/bash -l -o pipefail -c EFA_PC=$(find /opt/amazon/efa -path '*/pkgconfig/libfabric.pc' 2>/dev/null | head -n1) &&
      EFA_LIBFABRIC_RAW=$(cat "$EFA_PC" 2>/dev/null | grep '^Version:' | awk '{print $2}') &&
      EFA_LIBFABRIC_VER=$(echo "$EFA_LIBFABRIC_RAW" | grep -oE '^[0-9]+\.[0-9]+(\.[0-9]+)?') &&
      REF_VER=$(echo "${NIXL_LIBFABRIC_REF}" | sed 's/^v//') &&
      if [ -n "$EFA_LIBFABRIC_VER" ] &&
      [ -n "$REF_VER" ] &&
      [ "$(printf '%s\n' "$EFA_LIBFABRIC_VER" "$REF_VER" | sort -V | head -n1)" = "$EFA_LIBFABRIC_VER" ] &&
      [ "$EFA_LIBFABRIC_VER" != "$REF_VER" ]; then rm -rf /opt/amazon/efa &&
      cp -Pfr /tmp/libfabric_build /opt/amazon/efa &&
      sed -i 's|^prefix=.*|prefix=/opt/amazon/efa|' /opt/amazon/efa/lib/pkgconfig/libfabric.pc &&
      echo "/opt/amazon/efa/lib" > /etc/ld.so.conf.d/000_efa.conf &&
      rm -f /etc/ld.so.conf.d/efa.conf &&
      ldconfig &&
      echo "[aws] libfabric overlay: ${REF_VER} (overwrites EFA stock ${EFA_LIBFABRIC_RAW})"; else echo "[aws] libfabric overlay: skipped (EFA stock ${EFA_LIBFABRIC_RAW:-unknown} >= ${REF_VER})"; fi
    06/12/2026 8:26 PM UTC
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ARG
    NIXL_LIBFABRIC_REF=v2.5.1
    06/12/2026 8:26 PM UTC
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ENV
    EFA_VERSION=1.47.0
    06/12/2026 8:26 PM UTC
    sha256:fb69167348bde5267dfb8514e6411b747019c60b2b9a996669a5b3e52de0eb5dRUN
    MODELEXPRESS_VERSION=0.4.0 DEVICE=cuda PYTHON_VERSION=3.12 SITE_PACKAGES=/opt/dynamo/venv/lib/python3.12/site-packages ENABLE_KVBM=true ENABLE_GPU_MEMORY_SERVICE=true DYNAMO_COMMIT_SHA=919682da679aa699d5bca9c872f4c1d9a530bbc0 EFA_VERSION=1.47.0 /bin/bash -l -o pipefail -c mkdir -p /tmp/efa &&
      cd /tmp/efa &&
      curl --retry 3 --retry-delay 2 -fsSL -o aws-efa-installer-${EFA_VERSION}.tar.gz https://efa-installer.amazonaws.com/aws-efa-installer-${EFA_VERSION}.tar.gz &&
      tar -xf aws-efa-installer-${EFA_VERSION}.tar.gz &&
      cd aws-efa-installer &&
      apt-get update &&
      ./efa_installer.sh -y --skip-kmod --skip-limit-conf --no-verify &&
      rm -rf /tmp/efa &&
      rm -rf /opt/amazon/aws-ofi-nccl /etc/ld.so.conf.d/aws-ofi-nccl.conf &&
      ldconfig
    06/12/2026 8:26 PM UTC
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4USER
    root
    06/12/2026 8:25 PM UTC
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ARG
    EFA_VERSION=1.47.0
    06/12/2026 8:25 PM UTC
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4CMD
    06/12/2026 8:25 PM UTC
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ENTRYPOINT
    /opt/nvidia/nvidia_entrypoint.sh
    06/12/2026 8:25 PM UTC
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ENV
    VLLM_USE_FLASHINFER_SAMPLER=1
    06/12/2026 8:25 PM UTC
    ...