AWS Deep Learning Containers for vLLM-Omni 0.18.0 on SageMaker¶
AWS Deep Learning Containers for SageMaker are now available with vLLM-Omni 0.18.0.
Announcements¶
-
Initial release of vLLM-Omni containers for SageMaker
-
Includes ASGI routing middleware for /invocations dispatch via CustomAttributes
-
Built on Amazon Linux 2023 with Python 3.12 and CUDA 12.9
Core Packages¶
| Package | Version |
|---|---|
| vLLM | 0.18.0 |
| vLLM-Omni | 0.18.0 |
| PyTorch | 2.10.0 |
| TorchVision | 0.25.0 |
| TorchAudio | 2.10.0 |
| CUDA | 12.9.1 |
| FlashInfer | 0.6.6 |
| EFA | 1.47.0 |
Security Advisory¶
AWS recommends that customers monitor critical security updates in the AWS Security Bulletin.
Reference¶
Docker Image URIs¶
763104351884.dkr.ecr.us-west-2.amazonaws.com/vllm:omni-sagemaker-cuda-v1
public.ecr.aws/deep-learning-containers/vllm:omni-sagemaker-cuda-v1