# triple-mu/YOLOv8-TensorRT

YOLOv8 using TensorRT accelerate !

Repository: https://github.com/triple-mu/YOLOv8-TensorRT
Canonical: https://ross.abutalabs.com/products/yolov8-tensorrt
Language: Python
License: MIT
License Family: permissive
Topics: tensorrt, yolov8, onnx, deepstream, segment, jetson, detection, pose
Last push: 2026-08-14T12:26:23+00:00

## Health v2 (maintenance only)
Score: 75/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 97, release rhythm 35, longevity 95
- inputs: {"age_days": 1336, "days_push": 19, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1804, forks 299 (observed 2026-08-28T04:05:38.347187+00:00)

## What it is
A library for running YOLOv8 inference accelerated with NVIDIA TensorRT, supporting detection, segmentation, pose estimation, oriented bounding boxes, and classification from both Python and C++. It handles ONNX export, TensorRT engine building, and version-compatible deployment across TensorRT 8 through 11.

## Use cases
- accelerate yolov8 inference with tensorrt
- deploy object detection model on jetson
- run yolo segmentation in c++
- convert ultralytics model to tensorrt engine
- run pose estimation with gpu acceleration
- deploy yolo model with onnx and tensorrt
- high-speed object detection on nvidia gpu

## When to choose
- you need maximum inference speed for YOLOv8 on NVIDIA GPUs or Jetson devices
- you want both Python and C++ deployment paths sharing the same engines
- you need multiple YOLOv8 tasks (detect, segment, pose, OBB, classify) with one toolchain
- you must support multiple TensorRT versions without code changes

## When to avoid
- you are not using NVIDIA hardware, since TensorRT is NVIDIA-only
- you need a different model architecture than YOLOv8
- you want a simple CPU-only inference solution
- you need training or fine-tuning rather than inference

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, computer-vision, image-processing, gpu-computing, llm-inference
- domain: computer-vision, deep-learning, gpu-computing
- platform: python, cpp
- tags: yolov8, tensorrt, onnx, object-detection, instance-segmentation, pose-estimation, nvidia, deepstream, edge-deployment, linux, gpu, jetson

## Member repositories
- triple-mu/YOLOv8-TensorRT (main) score 75

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:38.347187+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 2026-08-30T03:21:50.193088+00:00, confidence not recorded.
  - readme: https://github.com/triple-mu/YOLOv8-TensorRT (fetched 2026-08-28T04:05:38.347187+00:00, sha fe159b7ce197)
- Data as of 2026-08-30T08:39:29.467469+00:00.