NVIDIA/TensorRT
NVIDIA SDK for optimizing and executing deep learning model inference on NVIDIA GPUs.
Not currently ranked — collecting fresh signals.
star history
TensorRT provides tools and libraries for optimizing and deploying neural network models for high-performance inference. It converts trained models into optimized inference engines, applies quantization and layer fusion optimizations, and executes them efficiently on NVIDIA GPU hardware. This open source repository includes plugins, ONNX parser components, and sample applications demonstrating inference capabilities.
Frequently asked
- What is NVIDIA/TensorRT?
- NVIDIA SDK for optimizing and executing deep learning model inference on NVIDIA GPUs.
- Is TensorRT open source?
- Yes — NVIDIA/TensorRT is open source, released under the Apache-2.0 license.
- What language is TensorRT written in?
- NVIDIA/TensorRT is primarily written in C++.
- How popular is TensorRT?
- NVIDIA/TensorRT has 13.2k stars on GitHub.
- Where can I find TensorRT?
- NVIDIA/TensorRT is on GitHub at https://github.com/NVIDIA/TensorRT.