ML engineering teams running distributed training and hyperparameter search jobs on shared GPU clusters.

Determined AI is an open source machine learning training platform providing distributed training, hyperparameter tuning, experiment tracking, and cluster resource management.

Self-hosted on-premises or in the cloud on Kubernetes or bare-metal GPU clusters, now maintained under HPE.

Used when a team needs to schedule and scale many training or fine-tuning jobs across a shared GPU cluster efficiently.

It bundles cluster scheduling, fault tolerance, and hyperparameter search so teams do not stitch these together from separate tools.

Supports PyTorch and TensorFlow training code, and runs on Kubernetes or Determined's own resource manager.

Open source, Apache 2.0 licensed, self-hosted, developed by Determined AI and now maintained by HPE following its 2021 acquisition.