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.
$npx -y skills add firecrawl/ai-research-skills --skill vllmInstalls into the current project.
Run `npx skills use "https://github.com/firecrawl/ai-research-skills" --skill "firecrawl/ai-research-skills/vllm"` and follow the generated skill instructions now. Read its complete output, redirecting it to a temporary file first if necessary. Resolve relative paths from the supporting-files directory it provides.
Use the skills in "https://github.com/firecrawl/ai-research-skills" that are relevant to the current task. Run `npx skills add "https://github.com/firecrawl/ai-research-skills"` and select the relevant skills, then follow their instructions.