meituan/YOLOv6

Health v2 · maintenance only

Adoption not part of the score

5895 stars · 1062 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

YOLOv6 is a single-stage object detection framework implemented in PyTorch, designed for industrial applications with a family of pretrained models (N/T/S/M/L) balancing accuracy and speed. It includes training, evaluation, quantization, and deployment tooling, plus variants for segmentation, face detection, and mobile/CPU (YOLOv6Lite).

Use cases

• train a custom YOLO object detection model on my own dataset
• run fast real-time object detection on GPU with pretrained weights
• deploy an object detection model with TensorRT for production
• detect faces in images with a YOLO model
• run object detection on mobile devices or CPU
• quantize a detection model for faster inference

When to choose

• you need a well-benchmarked YOLO detector with pretrained models and deployment support
• you want industrial-grade speed/accuracy tradeoffs including TensorRT and quantization
• you need variants like segmentation, face detection, or lightweight mobile models

When to avoid

• you need the latest actively developed detector (development has slowed since 2023)
• your project requires a permissive license (GPL-3.0)
• you need tasks beyond detection/segmentation such as pose estimation or tracking

Facets

framework · maturity maintenance

machine-learning deep-learning computer-vision image-processing computer-vision deep-learning machine-learning artificial-intelligence python cross-platform object-detection yolo pytorch model-training inference industrial-applications pretrained-models gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="meituan/YOLOv6")