lightly-ai/lightly-train

All-in-one training for vision models (YOLO, ViTs, RT-DETR, DINOv3): pretraining, fine-tuning, distillation.

60
/ 100
Established

LightlyTrain helps machine vision engineers and researchers create cutting-edge computer vision models faster and more efficiently. You provide your image datasets, and it trains state-of-the-art models for tasks like object detection or panoptic segmentation. This tool is for professionals building vision-powered applications for a variety of industries, including those deploying to edge devices.

1,359 stars. Actively maintained with 43 commits in the last 30 days.

Use this if you need to train high-performance object detection or segmentation models on your custom image data, particularly if you're working with self-supervised learning methods like DINOv2/v3 or deploying to resource-constrained devices.

Not ideal if you're looking for a general-purpose machine learning framework that handles tabular data, natural language processing, or other non-vision tasks.

computer-vision object-detection image-segmentation machine-learning-engineering edge-device-ai
No Package No Dependents
Maintenance 20 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 14 / 25

How are scores calculated?

Stars

1,359

Forks

63

Language

Python

License

AGPL-3.0

Last pushed

Mar 12, 2026

Commits (30d)

43

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/lightly-ai/lightly-train"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.