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 GPU memory. Supports OpenAI-compatible endpoints, quantization (GPTQ/AWQ/FP8), and tensor parallelism.
安装
npx skills add davila7/claude-code-templates --skill serving-llms-vllmnpx skills add davila7/claude-code-templates --skill serving-llms-vllm -a claude-code -gnpx skills add davila7/claude-code-templates --skill serving-llms-vllm -a codex -ggemini skills install https://github.com/davila7/claude-code-templates.git --path cli-tool/components/skills/ai-research/inference-serving-vllmgh skill install davila7/claude-code-templates serving-llms-vllmgit clone --depth 1 https://github.com/davila7/claude-code-templates /tmp/davila7__claude-code-templates
mkdir -p ~/.claude/skills ~/.agents/skills
cp -R /tmp/davila7__claude-code-templates/cli-tool/components/skills/ai-research/inference-serving-vllm ~/.claude/skills/serving-llms-vllm # Claude Code
cp -R /tmp/davila7__claude-code-templates/cli-tool/components/skills/ai-research/inference-serving-vllm ~/.agents/skills/serving-llms-vllm # Codex, Cursor, Gemini CLI, Copilot, OpenCode