Serves LLMs with high throughput using vLLM's PagedAttention and continuous batching.
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.This is the skill's description: the text your AI reads to decide when the skill applies.
orchestra-vllm/SKILL.md
---
name: serving-llms-vllm
description: 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.
version: 1.0.0
author: Orchestra Research
license: MIT
tags: [vLLM, Inference Serving, PagedAttention, Continuous Batching, High Throughput, Production, OpenAI API, Quantization, Tensor Parallelism]
dependencies: [vllm, torch, transformers]
---
# vLLM - High-Performance LLM Serving
## Quick start
vLLM achieves 24x higher throughput than standard transformers through PagedAttention (block-based KV cache) and continuous batching (mixing prefill/decode requests).
**Installation**:
```bash
pip install vllm
```
**Basic offline inference**:
```python
from vllm import LLM, SamplingParams
… load the full skill through Skill Me