# onnx/onnx-tensorrt

ONNX-TensorRT: TensorRT backend for ONNX

Repository: https://github.com/onnx/onnx-tensorrt
Canonical: https://ross.abutalabs.com/products/onnx-tensorrt
Language: C++
License: Apache-2.0
License Family: permissive
Topics: onnx, deep-learning, nvidia
Last push: 2026-08-03T17:24:44+00:00

## Health v2 (maintenance only)
Score: 92/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 95, release rhythm 84, longevity 100
- inputs: {"age_days": 3047, "days_push": 30, "days_rel": 30, "gap_med": 37, "n_releases_24m": 16}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3228, forks 548 (observed 2026-08-28T04:07:50.049103+00:00)

## What it is
A C++ parser library and backend that converts ONNX models into TensorRT engines for high-performance GPU inference. It is maintained by NVIDIA and supports dynamic shapes and a wide range of ONNX operators.

## Use cases
- run ONNX models with TensorRT for faster GPU inference
- convert ONNX models to TensorRT engines
- deploy deep learning models optimized for NVIDIA GPUs
- parse ONNX graphs into TensorRT networks
- benchmark ONNX model inference performance on NVIDIA hardware

## When to choose
- you have ONNX models and want TensorRT-optimized inference on NVIDIA GPUs
- you need dynamic shapes and full-dimension network support
- you want an NVIDIA-maintained parser integrated with the TensorRT ecosystem

## When to avoid
- you are not using NVIDIA GPUs or TensorRT
- your model uses ONNX operators not in the supported operator matrix
- you need a runtime-independent inference solution without a CUDA dependency

## Facets
- artifact type: library
- maturity: active
- function: parser, machine-learning, deep-learning, llm-inference
- domain: deep-learning, machine-learning, gpu-computing, developer-tools
- platform: cpp, python, windows
- tags: onnx, tensorrt, nvidia, model-inference, model-conversion, cuda, inference-optimization, linux, gpu, docker

## Member repositories
- onnx/onnx-tensorrt (main) score 92

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:50.049103+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:43:59.591453+00:00, confidence not recorded.
  - readme: https://github.com/onnx/onnx-tensorrt (fetched 2026-08-28T04:07:50.049103+00:00, sha a8a5eebcf85c)
- Data as of 2026-08-30T08:39:29.467469+00:00.