aws/deep-learning-containers resource

Health v2 · maintenance only

Flags: no_license

Adoption not part of the score

1186 stars · 555 forks observed · 2026-09-03

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

AWS Deep Learning Containers are pre-built, security-patched Docker images for running AI/ML workloads on AWS services like EC2, EKS, and SageMaker. The repository contains the build tooling and release pipelines for images covering PyTorch, TensorFlow, vLLM, SGLang, Ray, and more.

Use cases

• serve large language models on AWS with vLLM or SGLang
• run PyTorch or TensorFlow training on EC2 or SageMaker
• deploy ML models with Ray Serve or TensorFlow Serving
• transcribe and diarize audio with WhisperX
• serve quantized GGUF models with llama.cpp on CPU or GPU
• run multimodal TTS and image generation models
• get security-patched deep learning environments without setup

When to choose

• you run AI/ML workloads on AWS and want tested, pre-configured Docker images
• you need optimized LLM inference images for EC2, EKS, or SageMaker
• you want to avoid manually installing CUDA drivers and framework dependencies

When to avoid

• you deploy outside AWS and prefer building your own images
• you need a lightweight solution without Docker or GPU infrastructure
• you require a permissive open-source license (license is custom)

Facets

infra-config · maturity active

llm-inference machine-learning deep-learning container-runtime gpu-computing machine-learning deep-learning large-language-models cloud-computing gpu-computing cloud python docker-images aws sagemaker vllm pytorch tensorflow ec2 eks prebuilt-images containers docker gpu linux

Member repositories

For agents

markdown · JSON · MCP: product_card(name="aws/deep-learning-containers")