TensorRT: NVIDIA's inference optimizer and runtime

Fields

id

sources/src5-tensorrt-github

URL

https://github.com/NVIDIA/TensorRT

Publisher

NVIDIA

Kind

repo

Published

2019-01-01

Accessed

2026-10-04

notes

TensorRT provides optimization and runtime for deep learning inference. Supports FP16, INT8, FP8 quantization. Key for NVIDIA GPU inference. Used by TensorRT-LLM for optimized inference engines. Supports CUDA graphs, dynamic shapes, multi-stream execution.

Notes

TensorRT provides optimization and runtime for deep learning inference. Supports FP16, INT8, FP8 quantization. Key for NVIDIA GPU inference. Used by TensorRT-LLM for optimized inference engines. Supports CUDA graphs, dynamic shapes, multi-stream execution.

Cross-references

This record references no other records.