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
sha256:04cae7714dd0cbfca6254c6055b1c2dd0ea627e751167b9720fd6865926ec5c9COPY
amd64/ /legal/
07/27/2026 6:59 PM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ENTRYPOINT
07/27/2026 6:43 AM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ENV
DYNAMO_COMMIT_SHA=77ccc1ca9d9e34333071280fbed97ad87e38819a
07/27/2026 6:43 AM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ARG
DYNAMO_COMMIT_SHA=77ccc1ca9d9e34333071280fbed97ad87e38819a
07/27/2026 6:43 AM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4USER
dynamo
07/27/2026 6:43 AM UTC
sha256:39e9c1c77d4513116d409e1e7f990f191ac82bae6676d7b009c2155743ac3662RUN
PYTHON_VERSION=3.12 ENABLE_KVBM=true ENABLE_GPU_MEMORY_SERVICE=true VLLM_OMNI_REF=v0.25.0rc1 NIXL_REF=v1.1.0 CUDA_MAJOR=13 MODELEXPRESS_VERSION=0.4.0 SITE_PACKAGES=/usr/local/lib/python3.12/dist-packages sed '/^#\s/d' /opt/dynamo/launch_message.txt > /opt/dynamo/.launch_screen &&
  chmod 755 /opt/dynamo/.launch_screen &&
  echo 'cat /opt/dynamo/.launch_screen' >> /etc/bash.bashrc
07/27/2026 6:43 AM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4USER
root
07/27/2026 6:43 AM UTC
sha256:157bfac0b927b76f6fcb9571d0e05aca474b38390d6480dfaf0fda6ba74fcb09COPY
--chown=dynamo:0 lib /workspace/lib
07/27/2026 6:43 AM UTC
sha256:9bd74647c663b0f9d3e4b2a6796b4152e6be03521224eea6431d52252e2bbc66COPY
--chown=dynamo:0 --chmod=775 components/src/dynamo/vllm /workspace/components/src/dynamo/vllm
07/27/2026 6:43 AM UTC
sha256:300ebdb26412a3e03a81afa6cbc17bafd64c2edf5f82aa3a2f2b7aa99416e9c3COPY
--chown=dynamo:0 --chmod=775 components/src/dynamo/frontend /workspace/components/src/dynamo/frontend
07/27/2026 6:43 AM UTC
...