GKE Pod Snapshots Cut Model Load Times

Google published GKE Pod snapshot benchmarks showing startup latency cuts up to 89%, with a 70B model loading in 37 seconds and an 8B model in 15 seconds. The feature, generally available since May on clusters running 1.35.3-gke.1234000 or later, saves CPU and GPU memory state and restores it on demand, but requires gVisor sandboxing and leaves snapshot invalidation and rehydration to platform teams.