ModelRefs / Triton Inference Server — AI Glossary

Triton Inference Server — AI Glossary

NVIDIA's production inference server supporting multiple ML frameworks with GPU batching, model ensembles, and gRPC/HTTP APIs.

Overview

Triton (NVIDIA) serves models from PyTorch, TensorFlow, TensorRT, ONNX, and custom backends. Key features: dynamic batching, model ensembles (chain models into pipelines), concurrent model execution, model analyzer for performance tuning. Used by Microsoft Azure AI and major cloud providers for high-throughput LLM serving.

Reference details

Topicinfrastructure
Last reviewed2026-06-24

Continue your research

Use these connected ModelRefs sections to compare alternatives, inspect implementation paths, and review the evidence and governance boundaries relevant to Triton Inference Server — AI Glossary.

Frequently asked questions

What is Triton Inference Server?

NVIDIA's production inference server supporting multiple ML frameworks with GPU batching, model ensembles, and gRPC/HTTP APIs.

What concepts are related to Triton Inference Server?

Closely related concepts include ray serve, bentoml, vllm.