vLLM | NVIDIA NGC

NVIDIA

vLLM

Container

NVIDIA

vLLM

vLLM is a fast and easy-to-use library for LLM inference and serving. The NVIDIA vLLM NGC Container is optimized for GPU acceleration, and contains a validated set of libraries that enable and optimize GPU performance.

Layer	Label		Created
34e8e9b81a7ac75f18908a12d7df8acccea5582c54943367a3f6d4123d94b265	CONFIG	`Entrypoint /opt/nvidia/nvidia_entrypoint.sh; WorkingDir /workspace`	05/29/2026 8:37 AM UTC
27a86709070128803f3617828c4284f8bc9f563d87ef37da5387270276032b7d	ARG	`NVIDIA_BUILD_REF=43e8565002209616e3fc02778d7984465085cc9a`	05/29/2026 8:37 AM UTC
aafd56352cd5a7d3b348b929c25842b7d019d36852a47264ca74815a52ddd43a	LABEL	`com.nvidia.build.id=329071537`	05/29/2026 8:37 AM UTC
4cfd8788449c44fa9092d9e6614b435583c26d824e8c5de2f998b34861109aca	ENV	`NVIDIA_BUILD_ID=329071537`	05/29/2026 8:37 AM UTC
5972050f2814decf2d10374120586b8fde7142476a1bb2ecb7614dab24bf7c34	ARG	`NVIDIA_BUILD_ID=329071537`	05/29/2026 8:37 AM UTC
e8471a14521c4fc61c9d6ac308b5a47e9ad143516ab072377ebad9d825225601	ENV	`NVIDIA_VLLM_VERSION=26.05.post1`	05/29/2026 8:37 AM UTC
dd763ebc04208e08f3c7e90f852954b3729b6d132e14dcbff49a00cbe2ee9e08	ARG	`NVIDIA_VLLM_VERSION=26.05.post1`	05/29/2026 8:37 AM UTC
cd593a5afa1436c80835294b9ab2f3c97f86c7948e58b08f81059c5bb48cc23a	LABEL	`com.nvidia.vllm.version=0.21.0+2325b6f0`	05/29/2026 8:37 AM UTC
ed400f23c9d6810e1973a53a07838e86448a662b177852215b10f288c634239c	ENV	`PATH=/usr/local/lib/python3.12/dist-packages/torch_tensorrt/bin:/usr/local/cuda/bin:/usr/local/nixlbench/bin:/usr/local/nixl/bin:/usr/local/nvidia/bin:/usr/local/cuda/bin:/usr/local/mpi/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/local/ucx/bin:/opt/amazon/efa/bin:/opt/tensorrt/bin`	05/29/2026 8:37 AM UTC
a32f847fa3fc456c5d781b26bb6ecf0e154718768c1d5302a4a1667551d91e74	ENV	`NVIDIA_PRODUCT_NAME=vLLM`	05/29/2026 8:37 AM UTC