serving-llms-vllm

Serves LLMs with high throughput using vLLM's PagedAttention and continuous...

安装方式

默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。

用 Codex 或 Claude 帮你安装复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。

直接命令不会经过审查 Prompt；运行前请先检查来源。

npx skills add https://github.com/JackSmith111977/Hermes-Skill-View --skill serving-llms-vllm

命令会保持在同一行。复制前请横向滚动并检查完整内容。

想先保存到本地？可下载 SkillsMP 当前能够提供的文件。

决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。

SKILL.md

来源说明 · 只读预览

name	serving-llms-vllm
description	Serves LLMs with high throughput using vLLM's PagedAttention and continuous...
version	1.0.0
triggers	["serving llms vllm","serving-llms-vllm","vllm","模型部署","推理加速","LLM推理"]
author	Orchestra Research
license	MIT
dependencies	["vllm","torch","transformers"]
metadata	{"hermes":{"tags":["vLLM","Inference Serving","PagedAttention","Continuous Batching","High Throughput","Production","OpenAI API","Quantization","Tensor Parallelism"]}}
category	mlops

name	serving-llms-vllm
description	Serves LLMs with high throughput using vLLM's PagedAttention and continuous...
version	1.0.0
triggers	["serving llms vllm","serving-llms-vllm","vllm","模型部署","推理加速","LLM推理"]
author	Orchestra Research
license	MIT
dependencies	["vllm","torch","transformers"]
metadata	{"hermes":{"tags":["vLLM","Inference Serving","PagedAttention","Continuous Batching","High Throughput","Production","OpenAI API","Quantization","Tensor Parallelism"]}}
category	mlops