Optimizes LLM inference with NVIDIA TensorRT for maximum throughput and lowest latency. Use for production deployment on NVIDIA GPUs (A100/H100), when you need 10-100x faster inference than PyTorch, or for serving models with quantization (FP8/INT4), in-flight batching, and
Installs just this skill. Get the whole plugin for auto-invocation.
⚡ How it fires
How this skill gets triggered: by you, by Claude, or both.
Fires itselfClaude auto-loads it when your prompt matches the work.
You can call itInvoke it directly when you want it.
Slash command/inference-serving-tensorrt-llm
👁️ Context preview
The summary Claude sees to decide when to auto-load this skill.
Optimizes LLM inference with NVIDIA TensorRT for maximum throughput and lowest latency. Use for production deployment on NVIDIA GPUs (A100/H100), when you need 10-100x faster inference than PyTorch, or for serving models with quantization (FP8/INT4), in-flight batching, and
📊 Stats
Stars29,856
Forks3,189
LanguagePython
LicenseMIT
📦 Ships with claude-code-templates
</> SKILL.md
inference-serving-tensorrt-llm.SKILL.md
---name: tensorrt-llm
description: Optimizes LLM inference with NVIDIA TensorRT for maximum throughput and lowest latency. Use for production deployment on NVIDIA GPUs (A100/H100), when you need 10-100x faster inference than PyTorch, or for serving models with quantization (FP8/INT4), in-flight batching, and multi-GPU scaling.
version: 1.0.0
author: Orchestra Research
license: MIT
tags: [Inference Serving, TensorRT-LLM, NVIDIA, Inference Optimization, High Throughput, Low Latency, Production, FP8, INT4, In-Flight Batching, Multi-GPU]
dependencies: [tensorrt-llm, torch]
---# TensorRT-LLM
NVIDIA's open-source library for optimizing LLM inference with state-of-the-art performance on NVIDIA GPUs.
## When to use TensorRT-LLM
**Use TensorRT-LLM when:**
- Deploying on NVIDIA GPUs (A100, H100, GB200)
- Need maximum throughput (24,000+ tokens/sec on Llama 3)
- Require low latency for real-time applications
- Working with quantized models (FP8, INT4, FP4)
- Scaling across multiple GPUs or nodes
**Use vLLM instead when:**
- Need simpler setup and Python-first API
- Want PagedAttention without TensorRT compilation
- Working with AMD GPUs or non-NVIDIA hardware
