home/categories/machine-learning/melodic-software-claude-code-plugins-plugins-systems-design-skills-llm-serving-patterns-skill-md
machine-learningdata-ai

llm-serving-patterns

LLM inference infrastructure, serving frameworks (vLLM, TGI, TensorRT-LLM), quantization techniques, batching strategies, and streaming response patterns. Use when designing LLM serving infrastructure, optimizing inference latency, or scaling LLM deployments.

melodic-software
maintainer
melodic-software
Updated 1/19/2026
Stars
11
Forks
1
quick start

Installation and usage

LLM inference infrastructure, serving frameworks (vLLM, TGI, TensorRT-LLM), quantization techniques, batching strategies, and streaming response patterns. Use when designing LLM serving infrastructure, optimizing inference latency, or scaling LLM deployments.

Installation
$ install --globalskills.sh
Usage

Once installed, you can use this skill by running the following command in your terminal:

skills use llm-serving-patterns