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
    sglang-runtime-nightly
    Container
    NVIDIA
    NVIDIA
    sglang-runtime-nightly

    The Dynamo SGLang runtime nightly image is a containerized build of Dynamo + SGLang which serves as the base runtime environment for sglang based inference with Dynamo's distributed inference framework which is published every night.

  • LayerLabelCreated
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4CMD
    —
    		08/06/2026 9:04 AM UTC
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ENTRYPOINT
    /opt/nvidia/nvidia_entrypoint.sh
    		08/06/2026 9:04 AM UTC
    sha256:c1906bdb3d7e1ffa77b5a682caf7c88d5a8fb213cec530543ed5cc34d21cbb1fCOPY
    /legal /legal
    		08/06/2026 9:04 AM UTC
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ENV
    NVIDIA_DRIVER_CAPABILITIES=video,compute,utility
    		08/06/2026 9:04 AM UTC
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4CMD
    —
    		08/06/2026 9:04 AM UTC
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ENTRYPOINT
    /opt/nvidia/nvidia_entrypoint.sh
    		08/06/2026 9:04 AM UTC
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ENV
    DYNAMO_COMMIT_SHA=e07fa63e834df0d8040756ebfa5d2613542db2ad
    		08/06/2026 9:04 AM UTC
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ARG
    DYNAMO_COMMIT_SHA=e07fa63e834df0d8040756ebfa5d2613542db2ad
    		08/06/2026 9:04 AM UTC
    sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4USER
    dynamo
    		08/06/2026 9:04 AM UTC
    sha256:4f4fb700ef54461cfa02571ae0db9a0dc1e0cdb5577484a6d75e68dc38e8acc1RUN
    MODELEXPRESS_VERSION=0.5.0 ENABLE_GPU_MEMORY_SERVICE=true set -eux; remaining="$(find /usr /opt /workspace /sgl-workspace -xdev         \( -type f -o -type l \)         \( -name 'libx264*.so*' -o -name 'libx265*.so*'         -o -name 'libopenh264*.so*' -o -name 'libfdk-aac*.so*'         -o -name 'libfaac*.so*' -o -name 'libvo-aacenc*.so*'         -o -name 'libaacplus*.so*' \) -print)"; if [ -n "${remaining}" ]; then echo "ERROR: H.264/H.265/AAC codec libraries remain in the SGLang image:" >&2; echo "${remaining}" >&2; exit 1; fi; python3 -c 'import soundfile, torchaudio; from PIL import Image'
    		08/06/2026 9:04 AM UTC
    ...