vllm

Serves LLMs with high throughput using vLLM's PagedAttention and continuous batching. Use when deploying production LLM APIs, optimizing inference latency/throughput, or serving models with limited GP

by MesferAli· Repository·other
Also installable via skills CLI
npx skills add MesferAli/XCircle/.claude/skills/vllm

Source

Path:.claude/skills/vllm/SKILL.md(main)

Related in other

vllm | AgentArea Skills