菜单
Fast LLM inference engine. PagedAttention, continuous batching, tensor parallelism, speculative decoding, and prefix caching. OpenAI-compatible API server. Supports Llama, Mistral, Qwen, DeepSeek, and hundreds of models.
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。
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。
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/mkurman/zorai --skill 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。
基于 SOC 职业分类
正在显示 SKILL.md
| name | vllm |
| description | Fast LLM inference engine. PagedAttention, continuous batching, tensor parallelism, speculative decoding, and prefix caching. OpenAI-compatible API server. Supports Llama, Mistral, Qwen, DeepSeek, and hundreds of models. |
| tags | ["llm-serving","paged-attention","openai-compatible-server","high-throughput-inference","vllm"] |
vLLM is a high-throughput, memory-efficient LLM inference engine featuring PagedAttention (near-zero memory waste), continuous batching, tensor parallelism, speculative decoding, prefix caching, and an OpenAI-compatible API.
uv pip install vllm
from vllm import LLM, SamplingParams
llm = LLM(model="Qwen/Qwen2.5-1.5B-Instruct")
params = SamplingParams(temperature=0.7, top_p=0.9, max_tokens=512)
outputs = llm.generate(["What is the capital of France?"], params)
for o in outputs:
print(o.outputs[0].text)
vllm serve Qwen/Qwen2.5-1.5B-Instruct --port 8000
# OpenAI client:
curl http://localhost:8000/v1/chat/completions \
-H "Content-Type: application/json" \
-d '{"model": "Qwen/Qwen2.5-1.5B-Instruct", "messages": [{"role": "user", "content": "Hello!"}]}'
llm = LLM(model="meta-llama/Llama-3.1-8B", tensor_parallel_size=2)