lambda-labs-gpu-cloud

What problem does it solve?

This Skill provides access to powerful, on-demand GPU cloud instances, simplifying the setup and management of machine learning training and inference workloads.

Core Features & Use Cases

• GPU Instance Provisioning: Launch and manage various NVIDIA GPU instances (H100, A100, B200, etc.) with pre-installed ML stacks.
• Persistent Storage: Utilize attached filesystems to store datasets, models, and checkpoints across instance lifecycles.
• 1-Click Clusters: Deploy high-performance multi-node Slurm clusters for large-scale distributed training.
• Use Case: Train a large language model by launching an 8x H100 instance, attaching a persistent filesystem for your dataset and checkpoints, and running your distributed training script.

Quick Start

Use the lambda-labs-gpu-cloud skill to launch an instance with an H100 GPU and connect to it via SSH.