Library/pytorch-lightningForked

Lightning-AI/pytorch-lightning

pytorch-lightning

Pretrain, finetune ANY AI model of ANY size on 1 or 10,000+ GPUs with zero code changes.

View on GitHub↗Upstream Lightning-AI/pytorch-lightning↗

Builder

Lightning-AI

Lightning-AI • individual

Stars

30,986

Using upstream star count

Forks

3,699

Using upstream fork count

Open Issues

Activity Score

0/100

52 commits in 30d

Created

Mar 31, 2019

Project creation date

README Summary

PyTorch Lightning is a deep learning framework that enables researchers and practitioners to scale AI model training from single GPUs to thousands of GPUs without changing their code. It provides a high-level interface built on top of PyTorch that handles distributed training, logging, checkpointing, and other infrastructure concerns automatically. The framework supports pretraining and finetuning of models at any scale while maintaining code simplicity and reproducibility.

AI Dev Skills

Unmapped

Tags

Distributed Deep LearningGPU Computing and CUDAPyTorch Model DevelopmentTraining Loop OptimizationExperiment Tracking and MLOpsMixed Precision TrainingGradient AccumulationLearning Rate SchedulingModel CheckpointingHyperparameter OptimizationMulti-GPU TrainingCluster ComputingDeep Learning Framework DesignModel ParallelismData ParallelismMulti-GPU Model Fine-tuningFoundation Model TrainingHyperparameter Optimization at ScaleMultimodalLarge Language ModelsLarge Scale Model TrainingOn-premiseMulti-node ClustersDistributed AI TrainingCloud-based Deep Learning TrainingImageCloudAudioVideoEfficient Deep LearningTabularMLOps and Model Lifecycle ManagementSelf-hostedDistributed Deep Learning ExperimentsGPU ClustersTextScalable AI InfrastructureResearch Experiment ManagementAcademic Research InfrastructureProduction Model Training PipelinesPython

Taxonomy

AI Trends

Large Language Models Foundation Model Training Efficient Deep Learning MLOps and Model Lifecycle Management Distributed AI Training Scalable AI Infrastructure

Deployment Context

Cloud Self-hosted On-premise Multi-node Clusters GPU Clusters

Modalities

Text Image Audio Video Tabular Multimodal

Skill Areas

Use Cases

Large Scale Model Training Distributed Deep Learning Experiments Multi-GPU Model Fine-tuning Research Experiment Management Production Model Training Pipelines Hyperparameter Optimization at Scale Cloud-based Deep Learning Training Academic Research Infrastructure

Recent Activity

Updated 22 days ago

7 Days

30 Days

90 Days

105

Quality

production

Quality: high
Maturity: production

PM Skills

Developer Platform

Languages

Python100.0%

Timeline

Project created: Mar 31, 2019
Forked: Mar 23, 2026
Your last push: 22 days ago
Upstream last push: 6 days ago
Tracked since: Mar 22, 2026

Similar Repos

pgvector cosine similarity · $0

Loading…