tinyvision/DAMO-YOLO

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

3183 stars · 406 forks observed · 2026-08-28

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

DAMO-YOLO is a fast and accurate object detection framework built on PyTorch, featuring NAS-searched backbones, RepGFPN, a lightweight ZeroHead with AlignedOTA label assignment, and distillation enhancement. It provides pretrained models and end-to-end tools from training to ONNX/TensorRT deployment.

Use cases

• train a custom yolo object detection model
• deploy object detection with tensorrt or onnx
• find a fast real-time object detector alternative to yolov5
• run object detection on edge devices with low flops
• use pretrained models to boost downstream detection tasks
• quantize a detection model to int8

When to choose

• you need state-of-the-art speed/accuracy tradeoffs in the YOLO family
• you want a full train-to-deploy pipeline with TensorRT/ONNX export
• you need lightweight nano models for CPU or edge inference

When to avoid

• you need a general-purpose vision toolkit beyond detection
• you require frequent updates or broad community support
• your stack is TensorFlow or JAX based

Facets

library · maturity maintenance

machine-learning deep-learning computer-vision image-processing deep-learning computer-vision machine-learning artificial-intelligence python object-detection yolo pytorch onnx tensorrt nas model-deployment distillation gpu linux docker

Member repositories

For agents

markdown · JSON · MCP: product_card(name="tinyvision/DAMO-YOLO")