Tag: #llm-d
Kubernetes, DevOps and MLOps, written from things we ran and measured ourselves · 1 posts
When documentation still calls a metric name that vLLM deleted: How an llm-d docs PR was merged in 2 hours and 9 minutes
Three places in the llm-d documentation were calling metric names that current vLLM no longer exposes. The guide's vllm:gpuprefixcachehitrate was a V0 engine gauge that disappeared in vLLM 0.10, and the autoscaler design
2026-10-16 · 18 min read #오픈소스#오픈소스 기여#llm-d#vLLM#Prometheus