Özer Cem KELAHMETKELAHMET
Glossary · MLOps and AI infrastructure

What is Triton? Definition and meaning

A server that runs models from many frameworks with dynamic batching and concurrent execution on GPUs.

Definition

Triton: A server that runs models from many frameworks with dynamic batching and concurrent execution on GPUs.

In short: NVIDIA Triton Inference Server.

When it is used

High-throughput GPU inference.

Where it fits

Triton belongs to Inference and serving in the MLOps and AI infrastructure topic. These are the other terms in the same group:

  • vLLM: An engine that uses PagedAttention and continuous batching to serve language models efficiently.
  • SGLang: A serving framework with efficient prefix caching and a language for structured generation.
  • TensorRT: A compiler and runtime that optimizes neural networks for NVIDIA GPUs.
  • ONNX Runtime: A runtime that executes models in the ONNX format across hardware backends.
  • llama.cpp: A library that runs quantized language models on CPUs and a range of GPUs.

Learn more

The full guide to Triton explains how it works, when to use it and how it compares with the alternatives, with examples. The glossary lists every term on the site from A to Z.

Previous term: Trino · Next term: TRL

Last updated Nov 18, 2026
Type to search.
    ↑↓ navigate↵ open