FICHA · AUR

tensorrt

A platform for high-performance deep learning inference on NVIDIA hardware

  • Library
  • LIBRARY
  • Dependency only
official+codex · reviewed · May 29, 2026 description in en

Description

Neural-network models can be optimized and executed for high-performance inference on NVIDIA GPUs. This is useful for developers deploying computer vision, language, recommendation, or other deep-learning workloads where latency and throughput matter.

The platform depends on NVIDIA drivers, CUDA compatibility, and model conversion choices. Users should validate accuracy, supported layers, GPU memory use, and deployment constraints before using optimized engines in production.

Permissions

Permissions not analysed for this source yet.