Performance

Use these guides when throughput, memory, placement, or large-model training efficiency becomes the bottleneck.

Guide	What you get
Auto Placement	Auto-select the best placement for a workload.
Dynamic Scheduling	Dynamically schedule resources during training.
Profiling	System-level profiling of Ray worker processes.
5D Parallelism	Configure 5D parallelism for large models.
LoRA	Train with LoRA adapters.