OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Developer Tools & SDKs

YOLOV8 Tensorrt Cpp

YOLOV8 Tensorrt Cpp is an MIT-licensed C++ implementation that runs YOLOv8 GPU inference through TensorRT for object detection, semantic segmentation and body pose estimation.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

YOLOV8 Tensorrt Cpp — OpenViST project profile in AI Developer Tools & SDKs.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when you need a C++ TensorRT starting point for YOLOv8 GPU inference and can provide the required CUDA, TensorRT and OpenCV environment.

Who it is for

  • C++ developers implementing YOLOv8 computer-vision inference on a CUDA-capable system.
  • Developers exploring TensorRT-based workflows for detection, segmentation or pose estimation.

Documented uses

  • Run YOLOv8 inference on an image and save an annotated result.
  • Run inference from a webcam and display results in real time.
  • Benchmark an ONNX model with an input image.
  • Convert a PyTorch YOLOv8 model into ONNX for inference.
Before you choose

Requirements & limits

The project is available as a public GitHub repository under the MIT license.

  • The documented tested environments are Ubuntu 20.04 and 22.04; Windows is explicitly unsupported.
  • The setup requires CUDA, cuDNN, TensorRT 10 or newer, and OpenCV with CUDA support.
  • The project relies on the separate tensorrt-cpp-api library for inference.
  • The first run can take several minutes while TensorRT generates an optimized engine.
  • INT8 mode requires representative calibration data and trades accuracy for potential speed.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Convert a YOLOv8 PyTorch model to ONNX

Create an ONNX model for use with the project's TensorRT inference executables.

Before you start

  • A downloaded YOLOv8 PyTorch model.
  • Python pip and the Ultralytics package.
  • The repository cloned with its scripts directory.

Steps

  1. Install the package with the command below.

    Documented command
    pip3 install ultralytics

  2. From the repository's scripts/ directory, run the command below.

    Documented command
    python3 pytorch2onnx.py --pt_path

Expected outcome

The conversion script produces an ONNX model.

Guide limitations

  • When using another conversion script, disable its end2end option; this implementation performs bounding-box decoding and non-maximum suppression in C++.
  • The guide covers conversion only, not building or running the inference executables.
Documentation 1 ↗
Source-based practical guide

Build the C++ project

Compile the repository's C++ executables after installing and configuring its prerequisites.

Before you start

  • Ubuntu 20.04 or 22.04, the environments identified as tested in the README.
  • CUDA, cuDNN, build-essential, Python pip, CMake, OpenCV with CUDA support and TensorRT 10 or newer.
  • The repository cloned with --recursive and the TensorRT installation path set in CMakeLists.txt.

Steps

  1. Create a build directory with mkdir build.

  2. Enter it with cd build.

  3. Configure the project with cmake ...

  4. Compile with the command below.

    Documented command
    make -j

Expected outcome

The project is built with its documented executables available for subsequent use.

Guide limitations

  • The README states that Windows is not supported.
  • This is a build recipe; it does not establish that the build has been independently executed or tested.
Documentation 1 ↗
COMPARE OPTIONS

Compare options for Model inference and serving

These projects document a common task and share a project role. This is a source-based comparison, not an execution benchmark or a guarantee of interchangeability.

ProjectDocumented purposePlatformAccess and requirements
YOLOV8 Tensorrt CppRun YOLOv8 inference on an image and save an annotated result.Linux, ServerThe project is available as a public GitHub repository under the MIT license.
HyperposeCustomize pose-estimation architectures and datasets.Docker, Linux, macOS, Windows
Multi Task NlpDefine multiple tasks and their data and model settings in a YAML file.Linux, macOS, WindowsThe project is an open-source library under the Apache-2.0 license, distributed through its public GitHub repository.
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
14 October 2026
Original website title
cyrusbehr/YOLOv8-TensorRT-CPP · Official homepage title · retained in its source language
How we review sourcesSuggest a correction