The Curious Engineer

All blogs / AI / ML

Lightning AI

lightning.ai

Build, train and deploy your own specialized models. On-demand GPUs, clusters, AI coding agents and enterprise-grade infrastructure.

AI / MLPython

’21’22’23’24’25’26
Posts
175 from 26 May 2021 to 16 Oct 2026
Posting
weekly (35 in the last year)
Kind
company blog

2026 33

  1. NewReleasesLightning 2.1: Train Bigger, Better, FasterLightning AI is excited to announce the release of Lightning 2.1 ⚡.The theme this time around is “bigger, better, faster.”JP Hennessy· May 26, 2021 · 8 min read
  2. NewTrainingLightning Training IntroHow To Get Started Training With Lightning Lightning is a lightweight PyTorch wrapper for high-performance AI research, that reduces the boilerplate without limiting flexibility.JP Hennessy· May 26, 2021 · 1 min read
  3. NewTrainingAutomatic Batch Size Finder in LightningHow To Let Lightning Automatically Figure Out Batch Size Lightning is a lightweight PyTorch wrapper for high-performance AI research that reduces the boilerplate without limiting flexibility.JP Hennessy· May 27, 2021 · 1 min read
  4. NewTrainingAutomatic Learning Rate Finder in LightningHow To Let Lightning Find the Best Learning Rate Lightning is a lightweight PyTorch wrapper for high-performance AI research that reduces the boilerplate without limiting flexibility.JP Hennessy· May 28, 2021 · 1 min read
  5. NewTrainingExploding And Vanishing Gradients in LightningHow Lightning Identifies Exploding and Vanishing Gradients Lightning is a lightweight PyTorch wrapper for high-performance AI research that reduces the boilerplate without limiting flexibility.JP Hennessy· May 29, 2021 · 1 min read
  6. NewTrainingTruncated Back-propogation Through TimeHow Lightning Helps You With Long Sequences Lightning is a lightweight PyTorch wrapper for high-performance AI research that reduces the boilerplate without limiting flexibility.JP Hennessy· May 30, 2021 · 1 min read
  7. NewTrainingReload DataLoaders Every EpochHow Lightning Helps You Reload Your Data on Every Epoch Lightning is a lightweight PyTorch wrapper for high-performance AI research that reduces the boilerplate without limiting flexibility.JP Hennessy· June 1, 2021 · 1 min read
  8. NewTrainingAvoid Overfitting in Machine LearningUsing the EarlyStopping callback, Lightning will automatically stop training once a specific metric has stopped improving.JP Hennessy· June 1, 2021 · 1 min read
  9. NewTrainingGet Full Control of your Training Using CallbacksLightning has a technique to add arbitrary self-contained programs to your training which are called callbacks.JP Hennessy· June 1, 2021 · 1 min read
  10. NewOpinionWhy we open sourced our documentationGrid is a scalable machine learning platform that empowers researchers and data scientists to rapidly iterate on ideas without infrastructure hassles, while also enabling community-drivenJP Hennessy· June 24, 2021 · 1 min read
  11. NewCase StudiesManaging Machine Learning With Limited ResourcesIn the race to implement machine learning, businesses of all sizes look for more accurate predictions that will allow them to stay ahead and differentiate themselves from the competition.JP Hennessy· October 13, 2021 · 3 min read
  12. NewCase StudiesHow Neuroscientists Are Using AI To Understand BehaviorBefore discovering Grid, the team managed all their cloud experiments themselves on AWS instances using boto3 scripts and vanilla PyTorch.JP Hennessy· December 6, 2021 · 4 min read
  13. NewCase StudiesAvoid Building and Maintaining an MLOps Training EnvironmentUsing sophisticated machine learning techniques, the AutoDevTech platform accelerates a team’s efficiency by providing valuable insights into their development process in the context of industryJP Hennessy· January 25, 2022 ·
  14. NewCase StudiesHow Computer Vision Researchers Use Grid For Their Infrastructure RequirementsThe store of Clutterbot, a New Zealand-based startup that builds autonomous cleaning robotsJP Hennessy· February 18, 2022 · 3 min read
  15. NewNewsLightning AI Acquires TensorwerkA Letter From Luca Antiga, Co-Founder and CEO of Tensorwerk Since its inception, Tensorwerk has been about people.JP Hennessy· February 24, 2022 · 2 min read
  16. NewBuildIntroduction to PyTorchA StatQuest Tutorial PyTorch is one of the most popular tools for making Neural Networks. This StatQuest walks you through a simple example of how to use PyTorch one step at a time.JP Hennessy· April 21, 2022 · 1 min read
  17. NewCase StudiesSolving Complex Macroeconomic Problems With Machine LearningHow a team at The University of British Columbia was able to quickly train models and obtain results faster than ever beforeJP Hennessy· April 6, 2022 · 4 min read
  18. NewNewsWhat is GridThe Grid platform enables users to quickly iterate through the model development life cycle by managing the provisioning of machine learning infrastructure on the cloud.JP Hennessy· May 10, 2022 · 4 min read
  19. NewCase StudiesScaling Accelerated Drug DiscoverySyntheticGestalt, an AI startup based in London and Tokyo, is developing an automatic system to make valuable drug discoveries en masse.JP Hennessy· June 4, 2022 · 4 min read
  20. NewTrainingPerformance Notes Of PyTorch Support for M1 and M2 GPUsApple's M1/M2 chips, known for strong performance and energy efficiency, now support GPU acceleration in PyTorch, and while their GPU RAM usage is higher than CUDA GPUs, training with adjusted batchSebastian Raschka· June 6, 2022 · 1
  21. NewNewsGrid.ai Becomes Lightning AIGrid.ai is now Lightning AI. Here is what the rename means and why we are unifying the AI development lifecycle, along with our $40M funding round.JP Hennessy· June 27, 2022 · 5 min read
  22. NewBuildAn Introduction to Lightning AIWhat Lightning AI is, how it differs from Grid.ai, and how the open-source Lightning framework fits into the platform.JP Hennessy· June 27, 2022 · 5 min read
  23. NewBuildRecurrent Neural Networks (RNNs)A StatQuest Tutorial When you don’t always have the same amount of data, like when translating different sentences from one language to another, or making stock market predictions from differentJP Hennessy· July 10, 2022 · 1 min read
  24. NewOpinionStop Wasting Time on MLOpsThe Fragmented World of Machine Learning/MLOps At June’s inaugural Lightning Developer Conference, Founder and CEO William Falcon’s keynote explored the state of artificial intelligence, and how it’sJP Hennessy· July 15, 2022 · 8 min read
  25. NewBuildBuild with Lightning and Win! Introducing the Storm Chaser Challenge SeriesOur first Storm Chaser challenge series runs July 20 to August 3, 2022. Complete the tasks to win a $1,000 Visa gift card.JP Hennessy· July 20, 2022 · 1 min read
  26. NewBuildBuilding a GAN (Generative Adversarial Network)Train And Explore Generative Adversarial Networks (GANs) Interactively With Lightning AppsJP Hennessy· July 19, 2022 · 7 min read
  27. NewBuildBuild an Interactive Research Poster with Lightning AIAn Easy Way to Demonstrate Your Research A research poster is an attractive visual presentation of research that displays the information in a clear and concise manner.Aniket Maurya· July 21, 2022 · 4 min read
  28. NewBuildHow To Build a Super Resolution GAN Demo in Lightning AIBuilding and training your deep learning and machine learning models is great, but what do you do once you’ve worked out the kinks?Sebastian Raschka· July 25, 2022 · 2 min read
  29. NewBuildHow to Build Apps on Lightning AIAt June’s inaugural Lightning Developer Conference, Lightning AI’s Staff Product Manager Eden Afek explored how you can bring your machine learning models to life through AI-powered Lightning Apps.JP Hennessy· July 27, 2022 · 5 min read
  30. NewBuildUsing Redis in a Machine Learning ApplicationHow to use Redis in a database system for machine learning applicationsJP Hennessy· July 28, 2022 · 10 min read
  31. NewNewsLightning Welcomes Jatheen, Priya, Mike, and JP!Today, we’re excited to announce four senior executive hires who will enable us to advance our mission of accelerating an AI-powered world.JP Hennessy· August 10, 2022 · 5 min read
  32. NewData ScienceGenerate Synthetic Data with NVIDIA’s Omniverse Replicator Lightning App (beta)Start using the Omniverse Replicator Lightning App for free today. At SIGGRAPH , NVIDIA and Lightning AI debuted the beta of the Omniverse Replicator Lightning App .JP Hennessy· August 10, 2022 · 3 min read
  33. NewReleasesLightning 1.6: Habana Accelerator, Bagua Distributed, Fault-Tolerance Improvements6 with support for Intel’s Habana Accelerator, new efficient DDP strategy (Bagua), manual Fault-tolerance, and other stability and reliability changes.JP Hennessy· August 31, 2022 · 2 min read

2025 19

  1. NewReleasesLightning 1.7: Apple Silicon, Multi-GPU and moreWe’re excited to announce the release of PyTorch Lightning 1.7 ⚡️ (release notes!)JP Hennessy· August 31, 2022 · 7 min read
  2. NewNewsPyTorch Team at Lightning – Welcome Mike RuberryToday, I’m thrilled to share that Mike Ruberry, one of PyTorch’s core technical leads, will be joining Lightning.William Falcon· September 8, 2022 · 5 min read
  3. NewTrainingHow to Enable GPU-Accelerated Training on Apple Silicon in PyTorchthis tutorial shows you how to train models faster with Apple’s M1 or M2 chips.JP Hennessy· September 13, 2022 · 4 min read
  4. NewNewsPyTorch Moves to Linux Foundation, Chinese Corpus for NLP…Can a single policy control the movement of hundreds of different robots? Is it possible to make 3D mapping more accurate and cheaper at the same time?JP Hennessy· September 15, 2022 · 4 min read
  5. NewData ScienceTrain Machine Learning Models with Synthetic DataThe Lightning App built with NVIDIA Omniverse Replicator (try it for free today!) lets you generate synthetic data to enhance the training and performance of your AI models.JP Hennessy· September 20, 2022 · 4 min read
  6. NewTrainingHow to Enable Native Fully Sharded Data Parallel in PyTorchthis tutorial teaches you how to overcome hardware constraints when training large models using PyTorch’s new model sharding strategyJP Hennessy· September 22, 2022 · 2 min read
  7. NewNewsICLR Releases Submissions for 2023, The White House Pens an “AI Bill of Rights” and Meta’s Make-A-Video Excites CreativesThis week, the announcement of several text-to image, voice and video generation models are delighting researchers and creatives alike.JP Hennessy· October 5, 2022 · 5 min
  8. NewShipHow to Deploy Diffusion ModelsIn this tutorial, you’ll learn how to deploy diffusion models at scale and build a text-to-image generatorAniket Maurya· October 12, 2022 · 6 min read
  9. NewNewsZuckerberg’s Avatar Gains “Social Presence”, AI Learns How to Play Nice, and Building Diffusion Models Just Got Way EasierLightning is handing out the blueprints of Stable Diffusion and researchers are helping AI become better at cooperating with humans in strategic game play.JP Hennessy· Oct
  10. NewTrainingGuide to Distributed TrainingIn this tutorial, you’ll learn how to scale models and data across multiple devices using distributed training.JP Hennessy· October 26, 2022 · 5 min read
  11. NewNewsSteve Jobs is Resurrected, Meta Is Translating Unwritten Languages and AI is Running for OfficeResearchers continue to push the limits of diffusion models, Steve Jobs returns for a conversation with Joe Rogan and Jack Dorsey’s new decentralized social network is open for beta users.JP Henness
  12. NewNewsMeta AI Releases a Metagenomic Atlas, Space Force Is Back With AI, and DALL·E API Now Available in Public BetaMeta AI’s researchers are celebrating the release of an open atlas of 617 million metagenomic protein structures, AI is being deployed to forecast satellite failures in orbit, and dev
  13. NewReleasesLightning 1.8: Colossal-AI, Secrets for Apps, and moreWe’re excited to announce the release of Lightning 1.8 ⚡️ (release notes).JP Hennessy· November 1, 2022 · 5 min read
  14. NewNewsBumble Is Blocking 🍆 Pics, DeepMind Proposes Algorithm Distillation and Shutterstock Pairs Up with DALL-E 2RThis week, researchers are improving the audio quality of your streaming services and continuing to improve reinforcement learning algorithms.JP Hennessy· October 27, 2022 · 6 min read
  15. NewNewsColossal-AI Accelerates AIGC, Snowpark for Python Is GA and the New Beta Release of NVIDIA OmniverseResearchers are improving audio quality with real-time target sound extraction and complex music mixing.JP Hennessy· November 10, 2022 · 5 min read
  16. NewShipКак деплоить диффузионные (diffusion) моделиThis Russian-language blog post, completed by a member of the Lightning community, is a translation of the English originalAniket Maurya· November 11, 2022 · 6 min read
  17. NewShipDeploy OpenAI Whisper as a Cloud ProductA New ML Model, a New Startup Opportunity — give Echo, our free transcription app powered by OpenAI’s Whisper, a try!JP Hennessy· December 8, 2022 · 7 min read
  18. NewNewsOpenAI’s New Chatbot, Stable Diffusion 2.0, and the TLDR on the Top Papers at NeurIPS 2022Researchers are testing the limits of language models and datasets while diffusion models continue to evolve.0 on your own servers and have a surprisingly human conversation with OpenAI’s ChatGPT.JP Henn
  19. NewNewsNVIDIA and Microsoft Team Up To Build an AI Supercomputer, Meta Releases Galactica and Sony Patents a New ML SystemRobot dogs and LLMs are getting more capable. NVIDIA and Microsoft team up on an AI supercomputer, and Meta releases Galactica.JP Hennessy· November 17, 2022 · 4 min read

2023 66

  1. NewNewsSebastian Raschka Launches a Deep Learning Course, NFL Wants ML to Predict Injuries, and Researchers Find 4 Sub-Phenotypes of Long COVIDResearchers in France have identified the four sub-phenotypes of long COVID while researchers in the US just released the largest federated ML effort to date
  2. NewShipScale Model Serving with Dynamic Batching and AutoscalingHow to scale model serving in production with dynamic batching and autoscaling.Aniket Maurya· December 20, 2022 · 3 min read
  3. NewBuildHow to Build a Machine Learning Training and Deployment PipelineLearn how you can use Lightning to build a model training and deployment pipeline that is customizable, integrated with tools (like monitoring, data warehouses, and feature stores), and deployed inAniket Maurya· December 12, 202
  4. NewOpinionWhat WON’T Happen in AI in 2023For this special end-of-year edition of our newsletter, we are poopin’ on parades and kicking over crystal balls. Here is what we’re betting won’t happen in the coming year:JP Hennessy· December 29, 2022 · 4 min read
  5. NewNewsThe World’s First Robot Lawyer Is Here, Microsoft Hopes ChatGPT Will Make Bing Smarter and Scaling Model Serving in Prod Just Got EasierCan we we compress large language models for better performance? Can we rethink LLM processing? These are just a few of the questions researchers are asking
  6. NewNewsChatGPT Inspires Virtual AI “Wife”, DeepMind Introduces DreamerV3, and OpenAI Teases Plans for Premium ChatResearchers demonstrate surprising differences in causality-based localization vs. knowledge editing in language Models.JP Hennessy· January 12, 2023 · 4 min read
  7. NewTrainingTrain Scikit-learn Models on the CloudThe scale of machine learning datasets can often become large enough that migrating to a cloud provider for training, storing checkpoints, and logs becomes necessary.Aniket Maurya· January 12, 2023 · 5 min read
  8. NewBuild머신러닝 훈련 및 배포 파이프라인을 만드는 방법Lightning을 이용하여 어떻게 모델 훈련 및 배포 파이프라인을 만들 수 있는지 배웁니다. 해당 파이프라인은 커스터마이징 할 수 있고, monitoring, data warehouse, feature store 등 여러 기능과 통합되어 있으며, production에 배포되어 있습니다.Aniket Maurya· January 13, 2023 · 3 min read
  9. NewNewsPrinceton Student Creates Tool To Detect ChatGPT and an AI Bot Just Beat the Rocket League EliteThanks to researchers from DeepMind, you can now build your own transformer models from scratch!JP Hennessy· January 19, 2023 · 5 min read
  10. NewShipServe Stable Diffusion Three Times FasterServing large models in production with high concurrency the ability to serve multiple simultaneous inference requests and throughputthomas chaton· January 24, 2023 · 4 min read
  11. NewNewsOpen-Source Implementations of ChatGPT’s Training Algorithm Are Released and moreResearchers are hoping watermarking language models can help stop plagiarism while LAION and CarperAI are helping the public replicate ChatGPT on their own.JP Hennessy· January 27, 2023 · 5 min read
  12. NewData ScienceHow to Evaluate Machine Learning ModelsAs machine learning models continue to gain widespread adoption in industries ranging from healthcare and manufacturing to retail and logistics, it has become increasingly important that we evaluateJP Hennessy· February 2, 2023 · 4 min read
  13. NewNewsDiffusion Models Under Fire for Privacy Concerns and MoreResearchers are raising a flag about the privacy risks of the latest diffusion models and the AI detection model created by Princeton student is now GA.JP Hennessy· February 3, 2023 · 5 min read
  14. NewShipAccelerate Serving Stable Diffusion by Optimizing the Inference SchedulerLearn how to accelerate serving models with sequential inference steps, like Stable Diffusion.thomas chaton· February 8, 2023 · 8 min read
  15. NewNewsGoogle’s Bard Made a $100 Billion Mistake and Bing’s New Chat Capabilities Delights UsersThe battle of the chatbots continues with Bard’s reception peaking and falling in a single week while Bing’s new interface is causing a flurry of excitement.JP Hennessy· February 10, 2023 · 5 min read
  16. NewNewsAI Just Flew a Fighter Jet for 17 Hours, Commanded Super Mario Bros and MoreML twitter is in a flurry over ToolFormer, a model that is getting LLMs to use tools in a self-supervised manner.JP Hennessy· February 16, 2023 · 4 min read
  17. NewCase StudiesNixtla's AI models make better forecasts with less dataNixtla's new pretrained AI models let you forecast the future with almost no historical data.JP Hennessy· February 17, 2023 · 5 min read
  18. NewNewsWhy Google’s Researchers are Intentionally Poisoning Datasets and MoreCan intentionally poisoning web-scale training datasets be a practical approach to preventing risk? How can model predictions and their intended usage be better aligned?JP Hennessy· February 24, 2023 · 4 min read
  19. NewBuildObject Detection on the Cloud with YOLOv8Posted on February 28, 2023 by Aniket Maurya & Ayush Chaurasia - Community , TutorialsAniket Maurya· February 28, 2023 · 4 min read
  20. NewReleasesBe the First to Know: Lightning Open Source LaunchBe the First to Hear About What is Launching on March 15th! We’ve been hard at work implementing user feedback into the next release of PyTorch Lightning, and we’re excited to unveil it on March 15th.JP Hennessy· March 1, 2023 · 1 min read
  21. NewOptimizeUsing DeepSpeed to Optimize ModelsA discussion of various techniques that you can use to get the most out of your deep learning models.JP Hennessy· March 2, 2023 · 5 min read
  22. NewNewsMicrosoft’s Kosmos-1 Claims to Solve IQ Tests and OpenAI Introduces ChatGPT and Whisper APIsDevelopers can now integrate ChatGPT and Whisper models into their apps and products through OpenAI’s API. Microsoft researchers introduced Kosmos-1, a MLLM that claims to solve IQ tests.JP Hennessy· M
  23. NewOpinionAre Consistency Models the Next Big Thing?OpenAI is creating an entirely new class of generative models for images, and researchers are wondering how language models do in-context learning.JP Hennessy· March 10, 2023 · 4 min read
  24. NewReleasesPyTorch Lightning 2.0 FAQAnswers to common questions about PyTorch Lightning 2.0: what changed, what is simpler, and what Lightning Fabric adds for finer control.JP Hennessy· March 15, 2023 · 2 min read
  25. NewReleasesIntroducing PyTorch Lightning 2.0 and FabricToday, we’re introducing PyTorch Lightning 2.0—as well as Fabric, a new library—to continue unlocking unprecedented scale, collaboration, and iteration for researchers and developers.Luca Antiga· March 15, 2023 · 3 min read
  26. NewTrainingTraining Compiled PyTorch 2.0 with PyTorch LightningPyTorch 2.x aims to push the performance with model compilation further while maintaining its Pythonic nature and backward compatibility.Aniket Maurya· March 17, 2023 · 4 min read
  27. NewTrainingTraining Compiled PyTorch 2.0 with PyTorch LightningHow to use PyTorch 2.0 and train a compiled model with PyTorch Lightning 2.0. Find the full code used in this tutorial here.Aniket Maurya· March 17, 2023 · 4 min read
  28. NewBuildIntroduction to Lightning FabricLightning Fabric is a new, open-source library that allows you to quickly and easily scale models while maintaining full control over your training loop.JP Hennessy· March 20, 2023 · 8 min read
  29. NewOptimizeAccelerate PyTorch Code with FabricUse Lightning Fabric to train and accelerate a PyTorch model using mixed precision and distributed training.Aniket Maurya· March 22, 2023 · 4 min read
  30. NewOptimizeHow to Speed Up PyTorch Model TrainingIn this tutorial, we will finetune a DistilBERT model, a distilled version of BERT that is 40% smaller at almost identical predictive performanceSebastian Raschka· March 23, 2023 · 17 min read
  31. NewTrainingFinetuning LLMs on a Single GPU Using Gradient AccumulationLearn how to use gradient accumulation to train models with large batch sizes in order to work around hardware limitations when GPU memory is a concern.Sebastian Raschka· March 28, 2023 · 13 min read
  32. NewNewsHow to Contribute to LightningContributing to open-source projects is a great way to improve your coding skillsAniket Maurya· March 30, 2023 · 4 min read
  33. NewOpinionWhy We’re Building Lit-LLaMAWith Lit-LLaMA, our goal is to provide a clean, fully open, non harmful implementation of the LLaMA source code that can be rapidly developed alongside the community.Luca Antiga· April 4, 2023 · 5 min read
  34. NewTrainingAccelerating LLaMA with Fabric: A Comprehensive Guide to Training and Fine-Tuning LLaMAIn this tutorial, we will learn how to train and fine-tune LLaMA (Large Language Model Meta AI). Lit-LLaMA, a rewrite of LLaMA, can run inference on an 8 GB consumer GPU.Aniket Maurya· April 6, 2023 · 5
  35. NewTrainingUnderstanding Parameter-Efficient Finetuning of Large Language Models: From Prefix Tuning to LLaMA-AdaptersLearn how popular parameter-efficient finetuning methods for LLM work: prefix tuning, adapters, and LLaMA-Adapter.Sebastian Raschka· April 12, 2023 · 14 min read
  36. NewNewsHow to Contribute to Lit-GPT and Lit-LLaMALearn about the variety of ways to contribute to Lit-LLaMA, Lightning AI’s open-source rewrite of LLaMA.JP Hennessy· April 20, 2023 · 10 min read
  37. NewTrainingParameter-Efficient LLM Finetuning With Low-Rank Adaptation (LoRA)In the rapidly evolving field of AI, using large language models in an efficient and effective manner is becoming more and more important.Sebastian Raschka· April 26, 2023 · 15 min read
  38. NewNewsThe Ultimate Battle of Language Models: Lit-LLaMA vs GPT3.5 vs Bloom vs …Large Language Models (LLMs) are taking center stage, wowing everyone from tech giants to small business owners.Daniela Dapena· April 28, 2023 · 21 min read
  39. NewOptimizeAccelerating Large Language Models with Mixed-Precision TechniquesTraining and using large language models (LLMs) is expensive due to their large compute requirements and memory footprints.Sebastian Raschka· May 10, 2023 · 16 min read
  40. NewTrainingHow To Train Reinforcement Learning Model To Play Game Using Proximal Policy Optimization (PPO) AlgorithmLearn how to build and train a Reinforcement Learning model with PyTorch and Lightning Fabric.Aniket Maurya· May 14, 2023 · 8 min read
  41. NewTrainingHow To Finetune GPT Like Large Language Models on a Custom DatasetLearn how to finetune large language models (LLMs) on a custom dataset. We will be using Lit-GPT, an optimized collection of open-source LLMs for finetuning and inference.Aniket Maurya· May 19, 2023 · 6 min read
  42. NewNewsLang Segment Anything – Object Detection and Segmentation With Text PromptOur community member Luca Medeiros shows how he leveraged the Segment Anything Model from Meta AI and built his lang-segment-anything library for object detection and image segmentation based on textJP Hennessy· May 24,
  43. NewNewsFalcon – A guide to finetune and inferenceIn this blog you will learn about the latest open-source large language model Falcon.5 GB of GPU memory!Aniket Maurya· June 9, 2023 · 4 min read
  44. NewOpinionHow Military Veterans Can Get into AILearn how to gain the technical skills to break into AI after a successful military career.JP Hennessy· June 2, 2023 · 9 min read
  45. NewTrainingEfficient Initialization of Large ModelsOne of the challenges with LLMs is their cost and large memory footprint.1 release, we introduce new features that optimize all three stages of LLM usage: pretraining, finetuning, and inference!JP Hennessy· June 19, 2023 · 7 min read
  46. NewTrainingFinetuning Falcon LLMs More Efficiently With LoRA and AdaptersUsing parameter-efficient finetuning methods outlined in this article, it’s possible to finetune an open-source LLM like Falcon in 1 hour on a single GPU instead of a day on 6 GPUs.Sebastian Raschka· June 14, 2023 · 12 min read
  47. NewOptimizeFaster PyTorch Training by Reducing Peak Memory (combining backward pass + optimizer step)A generic four line code change can reduce peak memory by 20-40% when training large models.JP Hennessy· June 21, 2023 · 10 min read
  48. NewOptimizeTorchmetrics v1.0: Visualize model performance with 100+ metrics0, is now publicly available.0, we have also managed to implement over 100 metrics in torchmetrics. In this blogpost we go over some of the new features and improvements to the library.JP Hennessy· July 4, 2023 · 5 min read
  49. NewOptimizeOptimizing Memory Usage for Training LLMs and Vision Transformers in PyTorchPeak memory consumption is a common bottleneck when training deep learning models such as vision transformers and LLMs.Sebastian Raschka· July 2, 2023 · 16 min read
  50. NewOptimizeTorchMetrics — a short history to 100+ metricsAs a celebration of v1.0.0 of Torchmetrics being released, in this blogpost we will discuss the history of Torchmetrics and how the interface have evolved over time.JP Hennessy· July 10, 2023 · 5 min read
  51. NewNewsNeurIPS LLM Efficiency ChallengeFinetune Lit-GPT on 1 GPU in 1 day and get a chance to win a speaker slot at NeurIPS 2023.JP Hennessy· July 15, 2023 · 1 min read
  52. NewData ScienceTabular Classification with LightningLearn how to gain back research time by leveraging PyTorch Lightning for over 100 inbuilt methods, hooks, and flags that save you engineering hours on heavy lifts like distributed training inJP Hennessy· July 13, 2023 · 15 min read
  53. NewBuildHow to build a chatbot using open-source LLMs like Llama 2 and FalconIn this blog, we’ll learn how to build a chatbot using open-source LLMs. We will be using Lit-GPT and LangChain. Lit-GPT is an optimized collection of open-source LLMs for finetuning and inference.Aniket Maurya· July 22, 20
  54. NewNewsStability AI announced SDXL 1.0 – Trained using PyTorch LightningWe are excited to share that PyTorch Lightning continues to power Generative AI research for training large foundation model, SDXL from Stability AI, by simplifying distributed training for massiveJP Hennessy· July 27, 2023 · 1
  55. NewTrainingThe NeurIPS 2023 LLM Efficiency Challenge Starter GuideLarge language models (LLMs) offer one of the most interesting opportunities for developing more efficient training methods.Sebastian Raschka· August 10, 2023 · 22 min read
  56. NewBuildWhen to Use PyTorch Lightning or Lightning FabricDiscover how to use PyTorch Lightning or Lightning Fabric to pretrain your custom model checkpoints or finetune LLMs like CodeLLaMA and LLama 2.JP Hennessy· September 7, 2023 · 6 min read
  57. NewInferenceOptimizing LLMs from a Dataset PerspectiveDiscover new research directions to improve Large Language Models (LLMs) and learn how to enhance the performance of instruction-finetuned LLMs by concentrating on higher-quality data and exploringSebastian Raschka· September 14, 2023 · 13 min re
  58. NewOptimizeTorchmetrics v1.2What clustering is, why it’s a useful measure, and newly added metrics that can be used with code samples.JP Hennessy· September 22, 2023 · 5 min read
  59. NewInferenceWhat is QuantizationLearn how quantization reduces the memory footprint of models like Llama 2 as much as 4x!JP Hennessy· October 3, 2023 · 4 min read
  60. NewTrainingScaling Large (Language) Models with PyTorch LightningIn this blog, you will learn about techniques to train large models like Llama (or any LLM) and Stable Diffusion using distributed training strategy FSDP with PyTorch Lightning.Aniket Maurya· October 4, 2023 · 9 min read
  61. NewTrainingFinetuning LLMs with LoRA and QLoRA: Insights from Hundreds of ExperimentsLoRA is one of the most widely used, parameter-efficient finetuning techniques for training custom LLMs.Sebastian Raschka· October 12, 2023 · 16 min read
  62. NewBuildFrom PyTorch to PyTorch Lighting: Getting Started GuideManaging training loops, device placement, logging, and scaling in raw PyTorch quickly becomes tedious and error-prone as projects grow.JP Hennessy· October 13, 2023 · 4 min read
  63. NewBuildPyTorch Lightning for Dummies – A Tutorial and OverviewYou’ll learn to use PyTorch Lightning’s Core API features by completing an applied project to train a Language Transformer written in PyTorch on the WikiText2 dataset.JP Hennessy· October 17, 2023 · 12 min read
  64. NewTrainingDoubling Neural Network Finetuning Efficiency with 16-bit Precision TechniquesThis guide targets PyTorch model training, illustrating how you can adjust the floating point precision to drastically enhance training speed and halve memory consumption, all without compromisingSebastian Rasch
  65. NewBuildStep-By-Step Walk-Through of Pytorch LightningTraining deep learning models in raw PyTorch gets repetitive fast. This walkthrough trains a CIFAR-10 image classifier with PyTorch Lightning, plus loggers and callbacks.Aniket Maurya· October 19, 2023 · 8 min read
  66. NewInference4-Bit Quantization with Lightning FabricReaders will learn the basics of Lightning Fabric’s plugin for 4-bit quantization.JP Hennessy· November 6, 2023 · 2 min read

2022 45

  1. NewBuildQuickstart to Lightning FabricScaling raw PyTorch is painful: too much boilerplate, too many edge cases. PyTorch Lightning can feel too opinionated when you want full control.JP Hennessy· November 15, 2023 · 6 min read
  2. NewInference8-bit Quantization with Lightning FabricLearn how to shrink massive models like Llama 2 and SDXL with a single line of code using Lightning Fabric’s 8-bit quantization plugin.JP Hennessy· November 15, 2023 · 2 min read
  3. NewNewsLightning AI Joins AI Alliance To Advance Open, Safe, Responsible AICompany Joins International Community of AI Organizations, Led by IBM and Meta, That Includes Leading Startups, Academia, Research and Government OrganizationsJP Hennessy· December 7, 2023 · 2 min read
  4. NewCase StudiesReducing carbon emissions faster at the MIT Sustainable Design LabMIT’s sustainability lab helps cities plan and reduce their carbon footprint with AI.William Falcon· June 14, 2025 · 2 min read
  5. NewCase StudiesDecoding the brain by training 1000 AI models to predict movementAt the Paninski lab at Columbia University, researchers trained 1,000 AI models in just 2 weeks to accelerate research that connects animal movement to brain activity and brings us closer toWilliam Falcon· June 14, 2025
  6. NewCase StudiesHow GenreX brought AI music to market 70% fasterGenreX is using AI to reinvent how music is made for games, ads, and creators everywhere.William Falcon· June 14, 2025 · 3 min read
  7. NewTrainingB200 vs H100 for model trainingMost teams either overpay for B200s or underuse them. We benchmarked B200 vs H100 to give you a clear, no-fluff guide to making the right call for your next training run.Thomas Viehmann· July 3, 2025 · 6 min read
  8. NewCase StudiesSynthera AI Brings Generative Simulations to Finance 5 Months Faster with Lightning AIAn AI startup skipped 5 months of infrastructure work to focus on building what mattered.Amit Ponchick· July 11, 2025 · 3 min read
  9. NewCase StudiesInworld AI Cut Months Off Voice AI Development with Lightning AIInworld AI accelerated Voice AI while reducing costs and advancing time-to-market.Amit Ponchick· July 15, 2025 · 4 min read
  10. NewCase StudiesReticular Sped Up Drug Discovery Research From 6 to 3 Months with Lightning AIReticular, a biotech startup, accelerated drug discovery research by simplifying collaboration and speeding up model experimentationAmit Ponchick· July 15, 2025 · 3 min read
  11. NewCase StudiesSentinal4D Advances Cancer Research with Lightning AI, Achieving 2x Faster DevelopmentSentinal4D halved their oncology drug discovery time by optimizing training, simplifying compute access, and improving global team collaboration with Lightning AI.Amit Ponchick· July 15, 2025 · 5 min
  12. NewCase StudiesHow Columbia Labs Accelerates Neuroscience Research and Democratizes AI with Lightning AIDr. Matt Whiteway’s team at Columbia University designed and shared their pose estimation tool with labs everywhere, without installation, GPU setup, or onboarding headaches.Amit Ponchick· July 17
  13. NewCase StudiesHow NNAISENSE Built an AI-Powered Hedge-Fund And Cut Infrastructure Costs By 35%A hedge fund startup simplified infrastructure and focused on strategy, saving time and money with Lightning AI.Amit Ponchick· July 17, 2025 · 3 min read
  14. NewInferenceFrom Flask to vLLM: How Model Inference has evolved (2017-2025). from Flask chatbots to full AI systems. In this post, I share what’s changed in the model inference world, what’s worked for me, and how I think about serving models in 2025.William Falcon· August 3, 2025 · 19 min read
  15. NewCase StudiesHow this Team Won the LeRobot Hackathon by Folding Shirts and Reimagining RoboticsA three-person team taught a robot arm to fold t-shirts and sort medicine, finishing 1st in the U.S. and 2nd globally, all built on Lightning AI.Amit Ponchick· August 4, 2025 · 3 min read
  16. NewOpinionGPT 5 vs GPT OSS 20B vs GPT OSS 120BI put OpenAI’s newest models, GPT-5, GPT OSS 120B, and GPT OSS 20B, head-to-head on a real production fraud detection task. The goal was to find the fastest model that doesn’t sacrifice accuracy..William Falcon· August 9, 2025 · 12 min read
  17. NewCase StudiesSlalom Accelerates Model Development and Cuts Setup Time Using Lightning AISlalom sped up model development and reduced setup overhead, empowering a small team to deliver big results.Amit Ponchick· August 12, 2025 · 7 min read
  18. NewBuildBuild faster with Lightning: a developer-first GPU marketplaceAccess high performance GPUs with a full-featured platform. No tradeoffs.Amit Ponchick· August 14, 2025 · 5 min read
  19. NewNewsLightning AI Launches First GPU Marketplace Uniting NeoClouds and Hyperscalers, Cutting AI Costs by 70% For the first time, teams can access high-performance GPUs with a full-featured platform. No tradeoffs between flexibility, speed, or scale.Amit Ponchick· August 21, 2025 · 4 min read
  20. NewCase StudiesInfineon Builds a Language Model Small Enough to Run on a Microcontroller in 4 MonthsInfineon compressed 1.5-2 years of development into 4 months with Lightning AI.Amit Ponchick· October 21, 2025 · 5 min read
  21. NewOpinionFault-tolerant AI inference despite Cloudflare outageCloudflare is down again, and I just want to rant a bit about how to think about protecting against these scenarios if you are new to AI inference.William Falcon· November 18, 2025 · 4 min read
  22. NewBuildMigrating from Neptune to Lightning AI: Experiment Tracking with LitLoggerIntroduction to LitLogger With Neptune shutting down, many ML teams are facing the same question: "Where do we move our experiment tracking without rewriting our code or losing history?"Amit Ponchick· December 19, 2025
  23. NewNewsWhat Is OpenClaw (Formerly Clawdbot and Moltbot)?OpenClaw (formerly Clawdbot and Moltbot) is a viral autonomous AI agent that hit 100K stars and triggered Mac Mini sellouts.JP Hennessy· January 31, 2026 · 6 min read
  24. NewNewsWhy we merged Lightning AI and Voltage Park to build an AI-native cloudLightning AI and Voltage Park merge to build a vertically integrated, AI-native cloud - combining AI software and owned GPU infrastructure in one platform.William Falcon· January 21, 2026 · 4 min read
  25. NewBuildRunning OpenClaw safely - Mac Mini vs Cloud Laptop.OpenClaw is the first easy-to-use personal AI agent, but you want to run it safely. Here we compare running on a mac mini vs on the cloud.William Falcon· February 21, 2026 · 8 min read
  26. NewNewsLightning on Cisco AI Pods: Powering Enterprise AI End-to-EndRun enterprise AI end-to-end on Cisco AI Pods with Lightning. Train, deploy, and scale AI workloads across on-prem and hybrid cloud.Amit Ponchick· February 9, 2026 · 4 min read
  27. NewBuildWhat Is Autoresearch? The Open Source Project Running ML Experiments as you SleepAndrej Karpathy's autoresearch runs AI-driven ML experiments autonomously — no supervision needed. Learn how it works and run it on Lightning Studios.JP Hennessy· March 23, 2026 · 3 min read
  28. NewBuildWorking with the Lightning CLI to Manage Lightning StudiosHow to manage Lightning Studios, jobs, and GPU switching entirely from your terminal using the Lightning CLI.Joe Mannix· February 24, 2026 · 7 min read
  29. NewBuildWhat is ZeroClaw? A Rust-Powered OpenClaw Alternative What is ZeroClaw? Discover how this OpenClaw alternative replaces heavyweight AI frameworks, and leverages Rust for secure, local-first agent workflows.JP Hennessy· February 27, 2026 · 6 min read
  30. NewBuildRun Autonomous, Self-Evolving Agents Safely with Lightning + NVIDIA OpenShellLightning integrates with NVIDIA OpenShell by providing high-performance frontier open model endpoints, nabling agents to leverage our compute and models safely, efficiently, and at scale.Amit Ponchick· March 16, 20
  31. NewBuild5 Things to Do With MicroGPT to Get the Most Value Out of It5 real ways to get value from Andrej Karpathy's microGPT, from experiment tracking to transformer intuition.Frances Fedoriska· March 23, 2026 · 6 min read
  32. NewBuildComfyUI in the Cloud for Agencies and Creative TeamsLooking to run ComfyUI in the cloud without installing drivers or managing GPUs? Lightning lets you launch ComfyUI online in one click, with up to 80 free GPU hours and no subscription required.Amit Ponchick· April 3, 2026 · 5 min read
  33. NewBuildWhat Is an Agent Harness? What Engineers Need to UnderstandThe term "agent harness" is being thrown around on blogs, X, podcasts and in heated debates between household names.Frances Fedoriska· April 20, 2026 · 10 min read
  34. NewChangelogDeepSeek V4 Pro Day 0 SupportDeepSeek V4 Pro is now available on Lightning AI. Start building with it today using a single API call.Liana Koleva· April 23, 2026 · 1 min read
  35. NewBuildDeepSeek V4 Alters Everything We Knew About Price-Performance MathDeepSeek dropped two open-weight models engineers have been waiting on for months.Frances Fedoriska· April 26, 2026 · 6 min read
  36. NewNewsHow the PyTorch Lightning Community Discovered a Supply Chain Attack and Fixed it in 42 Minutes3. The attack targeted the distribution layer, not the source code repository.Frances Fedoriska· April 30, 2026 · 5 min read
  37. NewBuildBuilding Multimodal Agent Systems with NVIDIA Nemotron 3 Nano Omni on LightningBuild multimodal agent systems that can see, hear, and reason using NVIDIA Nemotron 3 Nano Omni on Lightning.Amit Ponchick· April 28, 2026 · 3 min read
  38. ChangelogNewChangelogModel API deprecation: DeepSeek V3.1, Llama 3.3 70B3 70B serverless Model APIs are scheduled to be retired at 5pm PT on May 15, at which time these models will be disabled and any requests made to them will return HTTP error code 400.Liana Koleva· May 5, 2026 · 1 min read
  39. NewReleasesFrontier reasoning at scale: NVIDIA Nemotron 3 Ultra for long-running agents on LightningDeploy Nemotron 3 Ultra, an open frontier reasoning model for long-running agents. Scale orchestration and reasoning with Lightning AI. No proprietary model lock-in.JP Hennessy· June 4, 2026 · 4 min r
  40. ChangelogNewChangelogModel API deprecation: Kimi K2.5, GLM 55 and GLM 5 serverless Model APIs are scheduled to be retired at 5pm PT on June 11, at which time these models will be disabled and any requests made to them will return HTTP error code 400.Liana Koleva· June 4, 2026 · 1 min read
  41. NewNewsWhy Top-Performing Teams Resort to Shadow AIShadow AI is when researchers build outside your approved AI infrastructure. Learn what causes it, why it is a growing enterprise risk, and how building on Lightning AI reduces your threat exposure.JP Hennessy· June 26, 2026 · 8 min read
  42. NewBuildFrom "Waiting for GPUs" to a Production App in a DayTherna Biosciences traded infrastructure bottlenecks for research velocity on Lightning AIJP Hennessy· June 30, 2026 · 6 min read
  43. ChangelogNewChangelogModel API deprecation: MiniMax M2.5, Nemotron 3 Ultra5 and Nemotron 3 Ultra serverless Model APIs are scheduled to be retired at 5pm PT on July 13, at which time these models will be disabled and any requests made to them will return HTTP error codeLiana Koleva· July 7, 2026 · 1
  44. NewReleasesIntroducing Lightning Cloud: instant H100 access on Lightning’s fully-integrated training and experimentation cloudLightning AI now owns and operates its H100 fleet end-to-end. Guaranteed on-demand and spot capacity with a 1 year commitment, built for enterprise AI teams.Neil Bhatt· July
  45. NewCase StudiesHow Lightning AI Built NVIDIA SuperPODs for Cursor’s RL TrainingWhen Cursor needed a partner to design, deploy, and operate the compute foundation for their next venture, they trusted Lightning AI to architect a customized NVIDIA HGX B200 SuperPOD that didn’tFrances Fedoriska· July 15

2021 12

  1. NewBuildOne Team, Full Stack: Fully Virtualized H100s, Zero CompromiseHow we built a performant AI-native cloud on owned H100 and B200 GPUs, with bare-metal NCCL speed inside fully managed VMs.Thomas Chaton· July 16, 2026 · 19 min read
  2. NewShipJuly Recap: What we shipped this monthWe covered a lot of ground with what we shipped in July 2026. If you missed our weekly recaps on LinkedIn and X, here’s what happened.Neil Bhatt· July 27, 2026 · 4 min read
  3. ChangelogNewChangelogModel API deprecation: Nemotron 3 Nano OmniThe Nemotron 3 Nano serverless Model API is scheduled to be retired at 5pm PT on August 19, at which time this model will be disabled, and any requests made to it will return HTTP error code 400.Liana Koleva· August 12, 2026 · 1 min rea
  4. NewBuildPyTorch Lightning checkpoints write up to 95% faster on Google Cloud Storage Rapid BucketsWe worked with Google Cloud to fix how PyTorch Lightning moves data between GPUs and object storage.Amit Ponchick· September 10, 2026 · 3 min read
  5. ChangelogNewChangelogModel API deprecation: GPT-OSS 20B, 120BThe GPT-OSS 20B and GPT-OSS 120B serverless Model APIs are scheduled to be retired at 5pm PT on September 15, at which time these models will be disabled and any requests made to them will returnLiana Koleva· September 9, 2026 · 1 min read
  6. NewShipAugust Recap: What We Shipped This MonthOur team covered a lot of ground between what we shipped and where we were in August. If you missed our weekly recaps on LinkedIn and X, here’s what happened.Neil Bhatt· August 31, 2026 · 3 min read
  7. ChangelogNewChangelogModel API deprecation: DeepSeek V4 ProThe DeepSeek V4 Pro serverless Model APIs is scheduled to be retired at 5pm PT on September 23, at which time this model will be disabled and any requests made to it will return HTTP error code 400.Liana Koleva· September 16, 2026 · 1 min re
  8. NewCase StudiesHow Pangram Saved 3 Engineer-Years Building AI Detectors with Lightning AIThe AI-detection startup scaled from 3 people to 25 without ever hiring a dedicated infrastructure team.Amit Ponchick· September 24, 2026 · 6 min read
  9. NewBuildSeptember Recap: What We Shipped This MonthSeptember was full speed ahead on what we shipped. If you missed our weekly recaps on LinkedIn and X, here's what happened.Neil Bhatt· September 30, 2026 · 2 min read
  10. NewsNewNewsOctober 2026 Security IncidentOctober 2026 Security Incident On October 15, 11:00 PM UTC we identified a security incident on the Lightning Platform involving unauthorized access to credentials.Neil Bhatt· October 16, 2026 · 1 min read
  11. NewBuildVibe Coding My Way to a Boltz-2 Structure Prediction AppIn this blog post, Amir Momen-Roknabadi walks through his efforts to "vibe code" his way into creating an app to run Boltz-2 and deploy it to the Lightning AI platform.Amir Momen-Roknabadi· January 1, 1 · 11 min read
  12. NewBuildBuilding Reliable Agent Systems with NVIDIA Nemotron 3 Super on LightningLearn how to build multi-agent systems with NVIDIA Nemotron 3 Super on Lightning.Amit Ponchick· January 1, 1 · 6 min read