skills.rest
Clay-HHK
Clay-HHKCOMMUNITY
Explore all skills by this author →
GitHub

lambda-labs-gpu-cloud

Provision dedicated GPU cloud instances and Slurm clusters for ML training.

2|Updated Apr 12, 2026
Search Tags:#gpu#ssh#slurm#persistent-storage#ml-training#gpu-cloud#lambda-labs
One-click install
npx skills add https://github.com/Clay-HHK/claude-config --skill lambda-labs-gpu-cloud-clay-hhk
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: lambda-labs-gpu-cloud
Source: https://github.com/Clay-HHK/claude-config/tree/main/skills/AI-research-SKILLs/09-infrastructure/lambda-labs
Command: npx skills add https://github.com/Clay-HHK/claude-config --skill lambda-labs-gpu-cloud-clay-hhk

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Provides a turnkey environment for provisioning dedicated GPU instances and high-performance clusters so ML teams can run long training jobs, fine-tune large models, and serve inference without managing low-level GPU drivers, networking, or persistent storage.

Core Features & Use Cases

  • Dedicated GPU instances: On-demand and reserved instances across a wide GPU lineup (B200, H100, GH200, A100, A10, A6000, V100) for development, fine-tuning, and production training.
  • Persistent filesystems: Shared NFS-style filesystems that persist checkpoints, datasets, and models across instance lifecycles to avoid repeated downloads.
  • 1-Click Clusters & Slurm: Rapidly launch 16–512 GPU Slurm clusters with InfiniBand for multi-node distributed training (DDP, FSDP, DeepSpeed).
  • Preinstalled Lambda Stack: CUDA, drivers, PyTorch, TensorFlow, NCCL and common tooling ready out of the box to reduce setup time.
  • API & CLI automation: Launch, list, and terminate instances programmatically and manage SSH keys and filesystems for reproducible workflows.
  • Use case: Launch an 8x H100 cluster, attach a persistent filesystem for datasets and checkpoints, and run distributed training with torchrun or Slurm.

Quick Start

Launch an H100 instance in your preferred region, attach a persistent filesystem named training-data, and provide your SSH key to start a training-ready VM.

Frequently Asked Questions about lambda-labs-gpu-cloud

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I launch a multi-node GPU cluster for distributed ML training?▼

To launch a multi-node GPU cluster for distributed ML training, use the API or CLI to rapidly provision 16–512 GPU Slurm clusters with InfiniBand interconnects. This provides a high-bandwidth environment for running DDP, FSDP, or DeepSpeed workflows across global regions.

Can I run PyTorch and TensorFlow without manually installing CUDA drivers?▼

Yes, you can run PyTorch and TensorFlow without manually installing CUDA drivers because instances boot with a preinstalled Lambda Stack. This environment includes CUDA, NCCL, drivers, and common ML tooling ready out of the box to reduce setup time.

What is the best way to persist datasets and model checkpoints across instance restarts?▼

The best way to persist datasets and model checkpoints across instance restarts is to attach shared NFS-style persistent filesystems. These filesystems retain your data across instance lifecycles, preventing repeated downloads when VMs are terminated and relaunched.

Does API and CLI automation support managing SSH keys and terminating instances programmatically?▼

Yes, API and CLI automation supports managing SSH keys and terminating instances programmatically. You can launch, list, and terminate instances while managing SSH keys and filesystems to ensure reproducible automated ML training workflows.

Which GPU instances are available for single-node development and fine-tuning?▼

Dedicated GPU instances available for development and fine-tuning include B200, H100, GH200, A100, A10, A6000, and V100. These on-demand and reserved instances support single-node development and production training across various global regions.

Related Skills

Explore More Software Engineering Skills→
eddieluengeddielueng
1

lambda-labs-gpu-cloud

Provision reserved Lambda Labs GPU instances with persistent filesystems and SSH access.

Community
Intermediate
afel6afel6

lambda-labs-gpu-cloud

Provision on-demand GPU cloud instances for training and inference.

Community
Advanced
sovr610sovr610

Distributed Scaling

Configure DDP and FSDP distributed training for BrainAI models across GPUs and nodes.

Community
Advanced
JuanMS20JuanMS20

lambda-labs-gpu-cloud

Provision and manage on-demand GPU cloud resources for ML training and inference.

Community
Advanced