pytorch/TensorRT
A PyTorch-to-TensorRT compiler that optimizes deep learning models for faster inference on NVIDIA GPUs.
Not currently ranked — collecting fresh signals.
star history
Torch-TensorRT compiles PyTorch, TorchScript, and FX models to TensorRT engines, enabling up to 5x inference speedup on NVIDIA GPUs. It integrates directly with PyTorch as a backend, allowing users to accelerate models with minimal code changes by leveraging TensorRT’s optimization and execution capabilities.
Frequently asked
- What is pytorch/TensorRT?
- A PyTorch-to-TensorRT compiler that optimizes deep learning models for faster inference on NVIDIA GPUs.
- Is TensorRT open source?
- Yes — pytorch/TensorRT is open source, released under the BSD-3-Clause license.
- What language is TensorRT written in?
- pytorch/TensorRT is primarily written in Python.
- How popular is TensorRT?
- pytorch/TensorRT has 3k stars on GitHub.
- Where can I find TensorRT?
- pytorch/TensorRT is on GitHub at https://github.com/pytorch/TensorRT.