# laugh12321/TensorRT-YOLO

🚀 Easier & Faster YOLO Deployment Toolkit for NVIDIA 🛠️

Repository: https://github.com/laugh12321/TensorRT-YOLO
Canonical: https://ross.abutalabs.com/products/tensorrt-yolo
Homepage: https://github.com/laugh12321/TensorRT-YOLO
Language: C++
License: GPL-3.0
License Family: copyleft
Topics: tensorrt, tensorrt-inference, tensorrt10, computer-vision, image-classification, instance-segmentation, object-detection, pose-estimation, rotated-object-detection, ultralytics, yolo, yolo11, yolov5, yolov8, yolo26
Last push: 2026-03-22T11:08:32+00:00

## Health v2 (maintenance only)
Score: 63/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 73, release rhythm 48, longevity 67
- inputs: {"age_days": 948, "days_push": 164, "days_rel": 265, "gap_med": 58, "n_releases_24m": 8}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1880, forks 194 (observed 2026-08-28T04:05:48.590696+00:00)

## What it is
A C++/Python deployment toolkit for running YOLO-family models (YOLOv3 through YOLO26) on NVIDIA GPUs using TensorRT, with custom plugins, CUDA kernels, and CUDA Graphs for accelerated inference. It supports object detection, instance segmentation, image classification, pose estimation, rotated object detection, and video analysis out of the box.

## Use cases
- deploy yolo models on nvidia gpu with tensorrt
- run yolov8 inference in c++
- convert ultralytics yolo model to tensorrt engine
- fast object detection inference with cuda graphs
- instance segmentation deployment on jetson
- pose estimation inference in python
- rotated object detection (obb) deployment
- video stream analysis with yolo

## When to choose
- you need maximum YOLO inference performance on NVIDIA hardware via TensorRT
- you want C++ and Python APIs with minimal third-party dependencies
- you need support for many YOLO variants and tasks (detect, segment, classify, pose, OBB) out of the box

## When to avoid
- you are not targeting NVIDIA GPUs or don't use TensorRT
- you need a non-YOLO model architecture
- you require a permissive license (project is GPL-3.0)

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, computer-vision, image-processing, gpu-computing, sdk
- domain: computer-vision, deep-learning, machine-learning, gpu-computing, image-processing
- platform: cpp, python
- tags: tensorrt, yolo, object-detection, instance-segmentation, pose-estimation, cuda, nvidia, inference-deployment, ultralytics, obb, linux, gpu

## Member repositories
- laugh12321/TensorRT-YOLO (main) score 63

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:48.590696+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 2026-08-30T03:14:00.668931+00:00, confidence not recorded.
  - readme: https://github.com/laugh12321/TensorRT-YOLO (fetched 2026-08-28T04:05:48.590696+00:00, sha e0392d94c8b5)
  - homepage: https://github.com/laugh12321/TensorRT-YOLO (fetched 2026-08-29T10:53:05.819241+00:00, sha 9a4cc5a0b466)
- Data as of 2026-08-30T08:39:29.467469+00:00.