Observability, top-tier GPUs, and commitment-based savings.

Serverless Metrics page

Time-series charts for pXX latencies, queue delay, throughput, and worker states for faster debugging and tuning.

H100 on RunPod

Add NVIDIA H100 instances for higher throughput and larger model footprints.

Savings Plans

Commitment-based discounts for predictable workloads to lower effective hourly rates.