Linaom1214/TensorRT-For-YOLO-Series

Health v2 · maintenance only

Flags: no_license

Adoption not part of the score

1162 stars · 190 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

A Python and C++ toolkit for running YOLO-series object detection models (YOLOv3 through YOLOv12, YOLOX) with NVIDIA TensorRT, including ONNX export and engine generation with optional end-to-end NMS plugin support. It supports FP32/FP16/INT8 post-training quantization for accelerated GPU inference.

Use cases

• deploy yolo models with tensorrt for fast inference
• convert yolov8 onnx to tensorrt engine
• run yolo object detection on nvidia gpu with c++
• quantize yolo model to int8 with tensorrt
• export ultralytics yolo to tensorrt with nms plugin
• speed up yolo inference using tensorrt

When to choose

• you need maximum inference speed for YOLO models on NVIDIA GPUs via TensorRT
• you want end-to-end engines with NMS baked in via the plugin
• you need INT8 post-training quantization for YOLO deployment

When to avoid

• you are not using NVIDIA GPUs or TensorRT
• you need a detector other than the YOLO family
• you want a maintained, licensed production library rather than a demo-style codebase

Facets

library · maturity active

machine-learning computer-vision image-processing gpu-computing llm-inference computer-vision deep-learning gpu-computing python cpp windows tensorrt yolo object-detection onnx nms-plugin int8-quantization model-deployment gpu linux

Member repositories

For agents

markdown · JSON · MCP: product_card(name="Linaom1214/TensorRT-For-YOLO-Series")