Documentation
Browse guides, references, and examples.
Platform
End-to-end tooling for building, training and deploying AI models on the Lightning cloud
Getting StartedGet up and running with Lightning AI in minutesCLIManage Studios, Jobs, Deployments, and VMs directly from your terminalSDKProgrammatic access to Lightning AI resources from PythonAI StudioPersistent, collaborative GPU cloud workspace with AI copilotsDeploymentsDeploy autoscaling, high-performance inference APIsClustersFrontier-grade clusters with high-performance networking and storageOrganizationsManage teams, permissions, and collaboration at scale
Open Source
Frameworks to simplify training, optimizing and deploying AI models
PyTorch LightningPretrain, finetune and deploy AI models on any hardware — focus on science, not engineeringFabricScale PyTorch code with minimal changes to distributed training across GPUs/TPUsTorchMetricsMachine learning metrics for distributed, scalable PyTorch applicationsThunderMake PyTorch models faster with a source-to-source compiler for PyTorchLitServeEasily serve AI models - Lightning fastLitAILLM router and minimal agent framework in one - Lightning fastLitDataTransform datasets at scale. Optimize data for fast AI model training.
Was this page helpful?
Thanks for the feedback.