The Complete Production SOP: How to Deploy and Run vLLM High-Throughput PagedAttention Inference in Autonomous AI Agents

---
name: "vllm-high-throughput-inference"
description: "Serve open-source LLMs at maximum throughput with PagedAttention and vLLM engine."
metadata:
  author: "vLLM Team"
  version: "1.0.0"
  runtime_compatibility:
    - "codex"
    - "claude-code"
    - "cursor"
    - "windsurf"
triggers:
  - "deploy vllm server"
  - "serve llama 3 with vllm"
  - "configure pagedattention vllm"
permissions:
  fs_read: true
  fs_write: "workspace"
  shell_exec: false
  net_access: false
---

import torch
import vllm

print(f"vLLM Version: {vllm.__version__}")
print(f"CUDA Available: {torch.cuda.is_available()}")
print(f"Device Count: {torch.cuda.device_count()}")
assert torch.cuda.is_available(), "CUDA device required for vLLM execution"

#!/usr/bin/env bash
set -euo pipefail

NUM_GPUS=$(nvidia-smi --query-gpu=name --format=csv,noheader | wc -l)

python3 -m vllm.entrypoints.openai.api_server \
    --model meta-llama/Llama-3.3-70B-Instruct \
    --tensor-parallel-size "${NUM_GPUS}" \
    --gpu-memory-utilization 0.95 \
    --max-model-len 8192 \
    --port 8000 \
    --host 0.0.0.0

from openai import OpenAI

client = OpenAI(base_url="http://localhost:8000/v1", api_key="token-unused")
response = client.chat.completions.create(
    model="meta-llama/Llama-3.3-70B-Instruct",
    messages=[{"role": "user", "content": "ping"}],
    max_tokens=5,
)
assert response.choices[0].message.content is not None
print("Healthcheck Passed: PagedAttention serving functional.")