In vLLM v0.23.0, vllm:kv_cache_usage_perc is a Prometheus gauge for KV-cache usage where 1 means 100 percent usage.

Metric

Applies when

Evidence

Does not establish

• how much free KV cache remains for a specific model
• whether a workload will preempt
• a universal alert threshold

Guards

• vllm_version = v0.23.0
• metric_name = vllm:kv_cache_usage_perc
• runtime_component = scheduler

Freshness

Pinned release-tag source snapshot; re-check if target phase-1 release changes or metrics API compatibility becomes customer-facing.