Lightning AI

Lightning AI is a cloud platform from the creators of PyTorch Lightning for building, training and deploying specialized AI models. It offers on demand GPUs and clusters, cloud development environments called Studios that connect to any local IDE or over SSH, AI coding agents, and tools across the model lifecycle: building datasets, prototyping, running parallel experiments, comparing runs, pretraining, post training, evaluation and inference. It also supports bring your own cloud, so teams can use AWS or GCP commitments inside a private VPC, and lists SOC2, HIPAA and GDPR compliance. It is aimed at PyTorch developers, researchers and AI teams. Free accounts include up to 80 free GPU hours, depending on the GPU chosen and with up to 30 credits to start, and one Studio running around the clock with a restart every 4 hours. Compute is billed by the second, with listed per GPU hourly rates of $0.55 for a T4, $0.79 for an L4, $2.14 for an L40S, $2.71 for an 80 GB A100, $4.68 for an H100 and $4.50 for an H200. Paid plans are Pro at $50 per month, or $20 per month billed annually, and Teams at $140 per user per month, or $119 per user per month billed annually. Free plans for academics are also listed, and Enterprise terms are available on request.