Use the correct vllm metric gpu_cache_usage_perc --> kv_cache_usage_perc#1905

k8s-ci-robot merged 1 commit into

kubernetes-sigs:mainkubernetes-sigs/gateway-api-inference-extension:mainfrom

ezrasilvera:vllm-metricezrasilvera/gateway-api-inference-extension:vllm-metricCopy head branch name to clipboard