Monitor, debug, and deploy

Track experiments, inspect logs, schedule production GPU workloads, and serve the models you train on AI Runtime.

• • Track runs, logs, and metrics
  • Track experiments, metrics, and artifacts with MLflow, and view logs and GPU resource usage.
• • Schedule GPU workloads
  • Deploy training code with Declarative Automation Bundles, schedule runs, and build multi-task GPU and CPU workflows.
• • Convert a workload to a bundle
  • Convert an AI Runtime workload YAML into Declarative Automation Bundles to manage it as a persistent job.
• • Configure bundle tasks
  • Package training code, select GPU compute and environments, and pass configuration into bundle tasks.
• • Serve trained models
  • Deploy models trained on AI Runtime to Model Serving for real-time and provisioned-throughput inference.