Which integrations are available for this server?

Supports discovering PyTorch Deep Learning Containers images, provides framework-specific best practices and optimization tips for training and inference. Supports discovering TensorFlow Deep Learning Containers images, provides framework-specific best practices and optimization tips for training and inference.

How do I use AWS Deep Learning Containers MCP Server?

1. Click on "Install Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@AWS Deep Learning Containers MCP Server What instance should I use for a 35GB model?" That's it! The server will respond to your query, and you can continue using it as needed. Here is a step-by-step guide with screenshots.

AWS Deep Learning Containers MCP Server

Official

by aws-samples

Overview Schema Related Servers Score Discussions

Python

Remote

AWS Deep Learning Containers MCP Server

A Model Context Protocol (MCP) server for AWS Deep Learning Containers (DLC) that provides tools for discovering, building, deploying, and troubleshooting DLC images.

Features

Dynamic DLC Image Discovery: Automatically fetches latest images from AWS DLC GitHub - always up-to-date
Image Building: Create custom Dockerfiles and build images based on DLC base images
Multi-Platform Deployment: Deploy to SageMaker, EC2, ECS, and EKS
Instance Recommendations: Get GPU instance recommendations based on model size and budget
Upgrade Support: Analyze upgrade paths and generate migration Dockerfiles
Troubleshooting: Diagnose common DLC issues with actionable solutions
Best Practices: Security, cost optimization, and deployment guidance
No AWS Credentials Required: Discovery tools work without AWS credentials

Related MCP server: AWS Documentation MCP Server

Quick Start

Option 1: Run with uv (Recommended)

# Clone the repo
git clone https://github.com/aws-samples/sample-dlc-mcp-server.git
cd sample-dlc-mcp-server

# Run directly with uv
uv run dlc-mcp-server

Option 2: Run with Docker

# Build the image
docker build -t dlc-mcp-server .

# Run the container
docker run -it --rm \
  -v ~/.aws:/root/.aws:ro \
  dlc-mcp-server

Option 3: Install locally

pip install -e .
dlc-mcp-server

MCP Client Configuration

For Amazon Q CLI

Add to ~/.aws/amazonq/mcp.json:

{
  "mcpServers": {
    "dlc-mcp-server": {
      "command": "uv",
      "args": ["--directory", "/path/to/sample-dlc-mcp-server", "run", "dlc-mcp-server"],
      "timeout": 120000
    }
  }
}

For Kiro

Add to .kiro/settings/mcp.json:

{
  "mcpServers": {
    "dlc-mcp-server": {
      "command": "uv",
      "args": ["--directory", "/path/to/sample-dlc-mcp-server", "run", "dlc-mcp-server"],
      "timeout": 120000
    }
  }
}

Using Docker

{
  "mcpServers": {
    "dlc-mcp-server": {
      "command": "docker",
      "args": ["run", "-i", "--rm", "-v", "~/.aws:/root/.aws:ro", "dlc-mcp-server"],
      "timeout": 120000
    }
  }
}

Available Tools

DLC Discovery

Tool	Description
`search_dlc_images`	Search DLC images by framework, version, accelerator, platform
`get_dlc_recommendation`	Get image recommendations based on model type and size
`list_dlc_frameworks`	List all available frameworks with versions
`get_llm_serving_options`	Compare vLLM, SGLang, DJL, NeuronX options
`compare_dlc_images`	Side-by-side image comparison
`refresh_dlc_catalog`	Force refresh image catalog from GitHub

Image Building

Tool	Description
`create_custom_dockerfile`	Generate Dockerfile with custom packages
`build_custom_dlc_image`	Build and optionally push to ECR

Deployment

Tool	Description
`deploy_to_sagemaker`	Deploy to SageMaker endpoint
`deploy_to_ec2`	Launch EC2 instance with DLC
`deploy_to_ecs`	Deploy to ECS cluster
`deploy_to_eks`	Deploy to EKS cluster
`get_sagemaker_endpoint_status`	Check endpoint status

Instance Advisor

Tool	Description
`get_instance_recommendation`	GPU instance recommendations by model size
`list_gpu_instances`	List available GPU instances with pricing
`estimate_training_cost`	Estimate training job costs

Troubleshooting

Tool	Description
`analyze_dlc_error`	Analyze error logs with root cause analysis
`diagnose_common_issues`	Diagnose common DLC problems
`get_framework_compatibility_info`	Check framework version compatibility

Best Practices

Tool	Description
`get_security_best_practices`	Security guidelines
`get_cost_optimization_tips`	Cost reduction strategies
`get_deployment_best_practices`	Platform-specific guidance
`get_framework_specific_best_practices`	Framework optimization tips

Supported Frameworks

Framework	Latest Version	Use Cases
PyTorch	2.9.0	Training, Inference
TensorFlow	2.19.0	Training, Inference
vLLM	0.15.1	LLM Inference
SGLang	0.5.8	LLM Inference
HuggingFace PyTorch	2.6.0	NLP Training/Inference
AutoGluon	1.5.0	AutoML
DJL	0.36.0	Large Model Inference
PyTorch NeuronX	2.9.0	Trainium/Inferentia

Example Usage

Find vLLM images

Search for vLLM images for SageMaker inference

Deploy LLM to SageMaker

Deploy Qwen2.5-32B using vLLM on SageMaker with the right instance type

Get instance recommendations

What instance should I use for a 35GB model?

Troubleshoot errors

Help me fix this CUDA out of memory error: [paste error]

Configuration

Environment variables:

Variable	Description	Default
`ALLOW_WRITE`	Enable build/deploy operations	`false`
`ALLOW_SENSITIVE_DATA`	Enable detailed logs access	`false`
`FASTMCP_LOG_LEVEL`	Logging level	`ERROR`
`FASTMCP_LOG_FILE`	Log file path	None

Development

# Install dev dependencies
pip install -e ".[dev]"

# Run tests
python -m pytest tests/ -v

# Run linting
ruff check .

See DEVELOPMENT.md for more details.

License

This library is licensed under the MIT-0 License.

This server cannot be installed

license - permissive license

quality - not tested

maintenance

How are these scores calculated?

Maintenance

–Maintainers

–Response time

–Release cycle

–Releases (12mo)

Commit activity

Resources

GitHub Repository

Need Help?

Related Servers

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Latest Blog Posts

Your AI Chatbot Just Exposed Your CEO's Salary to an Intern
By Om-Shree-0709 on July 2, 2026.
Agent Identity
MCP Security
OAuth Delegation
Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)
By Om-Shree-0709 on June 30, 2026.
Agentic Ai
Prompt Injection
WebAssembly
Lightport: Open-Sourcing Glama's AI Gateway
By punkpeye on April 27, 2026.
OpenAI
open source

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/aws-samples/sample-dlc-mcp-server'

If you have feedback or need assistance with the MCP directory API, please join our Discord server