vllm-configuration

Configure and verify vLLM deployments in air-gapped environments.

5|1|Updated Apr 19, 2026
One-click install
npx skills add https://github.com/air-gapped/skills --skill vllm-configuration
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: vllm-configuration
Source: https://github.com/air-gapped/skills/tree/main/.claude/skills/vllm-configuration
Command: npx skills add https://github.com/air-gapped/skills --skill vllm-configuration

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Operators deploying vLLM in production often battle layered configuration surfaces (CLI flags, YAML files, and environment variables) and air-gap constraints, leading to misconfigurations and startup failures.

Core Features & Use Cases

  • Clear precedence rules across CLI > YAML > defaults, plus guidance for multi-source configs.
  • Nested config sections (ModelConfig, LoadConfig, CacheConfig, etc.) with validation and troubleshooting references.
  • Air-gap workflows, offline mode setup, and verification steps to ensure reliable deployments.

Quick Start

Create a minimal prod.yaml with your model path and key settings, then run vllm serve --config prod.yaml to launch the server.

Frequently Asked Questions about vllm-configuration

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I configure vLLM for air-gapped deployments?▼

vLLM configuration for air-gapped deployments involves creating a minimal prod.yaml with your offline model path and key settings, then launching the server using vllm serve --config prod.yaml to ensure reliable startup.

What is the precedence order for vLLM configuration sources?▼

vLLM configuration precedence strictly follows CLI arguments over YAML config files, which override default values, ensuring multi-source configs resolve predictably across multi-model setups and production clusters.

How do I troubleshoot vLLM startup failures caused by environment variables?▼

Troubleshoot vLLM startup failures by checking environment-variable gates against nested config sections like ModelConfig and LoadConfig, using the provided air-gap workflows and offline-mode guidelines to verify settings.

Can I use YAML files to manage multi-model vLLM cluster configurations?▼

Yes, YAML files can manage multi-model vLLM cluster configurations by defining per-section schemas such as CacheConfig and LoadConfig, with CLI overrides available to handle production-grade deployments.

Why does my vLLM offline mode setup fail during configuration verification?▼

vLLM offline mode setup fails during verification when layered configuration surfaces conflict; validate your CLI flags and YAML schemas against the documented air-gap workflows to resolve misconfigurations.