skills.rest
JamesFincher
JamesFincherCOMMUNITY
Explore all skills by this author →
GitHub

lambda-labs-gpu-cloud

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

Updated May 4, 2026
Search Tags:#gpu#pytorch#persistent storage#cloud provisioning#distributed training#ml training#ssh access
One-click install
npx skills add https://github.com/JamesFincher/gengar --skill lambda-labs-gpu-cloud-jamesfincher
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: lambda-labs-gpu-cloud
Source: https://github.com/JamesFincher/gengar/tree/main/optional-skills/mlops/lambda-labs
Command: npx skills add https://github.com/JamesFincher/gengar --skill lambda-labs-gpu-cloud-jamesfincher

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires lambda-cloud-client>=1.0.0, and includes references (resource) and assets (resource) components.

What problem does it solve?

This Skill solves the problem of running reliable ML training and inference workloads on reserved, on-demand GPU instances with straightforward SSH access and persistent storage.

Core Features & Use Cases

  • GPU cloud instance launch and SSH access: Choose GPU type and region, start an instance, and connect via SSH for hands-on experimentation or production jobs.
  • Persistent filesystems for data and checkpoints: Keep datasets, checkpoints, and outputs across restarts by mounting Lambda filesystems at a stable path.
  • Distributed and multi-node training support: Prepare for single-node multi-GPU and multi-node (Slurm/torchrun) training using pre-installed high-performance ML stack components.

Use case: You need an H100 cluster-style environment to fine-tune a large model and reliably resume from checkpoints stored on a persistent filesystem.

Quick Start

Tell the agent to set up an H100 instance in your chosen region, mount a filesystem for persistent checkpoints, then verify CUDA and PyTorch are working by running a GPU check and a CUDA availability check.

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 provision a GPU cloud instance for ML training with SSH access?▼

To provision a GPU cloud instance for ML training, select your desired GPU type and region, start the instance, and connect directly using SSH. This provides immediate, hands-on access for running development and training jobs.

Can I keep my datasets and model checkpoints across instance restarts?▼

Yes, you can keep datasets and checkpoints across restarts by mounting persistent filesystems. Mounting a Lambda filesystem at a stable path ensures your training outputs and data remain available even after the underlying instance is stopped or restarted.

Does this support multi-node distributed training with PyTorch and torchrun?▼

Yes, distributed training with PyTorch and torchrun is supported. The environment prepares single-node multi-GPU and multi-node setups by providing pre-installed high-performance ML stack components required for Slurm or torchrun execution across reserved instances.

How do I verify CUDA and PyTorch are working after launching a cloud GPU?▼

Verify CUDA and PyTorch are working on your cloud GPU by running nvidia-smi to check the driver and executing torch.cuda checks. These commands confirm the hardware is active and the framework correctly detects the available accelerators.

Do I need Lambda Labs credentials to set up a reserved H100 instance?▼

Yes, you need Lambda Labs credentials to set up a reserved H100 instance. The provisioning workflow requires authenticating with Lambda Labs via the API or console to select instance types, configure SSH keys, and mount optional filesystems.

Related Skills

Explore More Software Engineering Skills→
rawvegrawveg
40

vastai-api

Manage Vast.ai GPU instances and resources via REST API and CLI.

Community
Advanced
qcmuuqcmuu

lambda-labs-gpu-cloud

Provision Lambda Labs GPU instances and Slurm clusters for ML workloads.

Community
Intermediate
MesferAliMesferAli

lambda-labs-gpu-cloud

Provision on-demand NVIDIA GPU cloud instances with SSH access and persistent filesystems.

Community
Advanced
GarrettRoiGarrettRoi

lambda-labs-gpu-cloud

Provision on-demand NVIDIA GPU cloud instances and Slurm clusters for ML workloads.

Community
Advanced