{"repo":"lightly-ai/lightly-train","free":true,"listed":false,"github":"https://github.com/lightly-ai/lightly-train","clone":"git clone https://github.com/lightly-ai/lightly-train.git","description":"All-in-one training for vision models (YOLO, ViTs, RT-DETR, DINOv3): pretraining, fine-tuning, distillation.","language":"Python","stars":1645,"topics":["computer-vision","contrastive-learning","deep-learning","distillation","embeddings","pretrained-models","python","pytorch","self-supervised","self-supervised-learning"],"license":"AGPL-3.0","category":"machine-learning","readme_excerpt":"SOTA Pretraining, Fine-tuning and Distillation Train Better Models, Faster LightlyTrain is the leading framework for transforming your data into state-of-the-art computer vision models. It covers the entire model development lifecycle from pretraining DINOv2/v3 vision foundation models on your unlabeled data to fine-tuning transformer and YOLO models on detection and segmentation tasks for edge deployment. Struggling to get good results with pre-training? Talk to one of our experts Contact us Using LightlyTrain at work, in production, on the edge, or to build proprietary models? You likely need a Commercial License. Contact us to request a license for commercial use. Also check out LightlyStudio to easily visualize your annotations and predictions. News - \\[0.17.0\\] - 2026-07-28: LTDETRv2 for instance segmentation: Train state-of-the-art LTDETRv2 instance segmentation models with ECViT backbones from EdgeCrafter, matching the accuracy of the original ECSeg implementation while being 10-20% faster! ONNX and TensorRT export is also out-of-the-box! - \\[0.16.0\\] - 2026-06-25: ⚡ Upgraded LTDETRv2 for object detection: Following the success of LTDETR, LightlyTrain's DETR model, we release LTDETRv2 with significant architectural and performance improvements! It supports using ECViT backbones from EdgeCrafter and ONNX/TensorRT export for faster inference! - \\[0.15.0\\] - 2026-04-14: 🔎 Distillationv3: Better generalizing distillation method that performs equally well across dense and ","default_branch":null,"files":null,"tree":[],"storefront":"/r/lightly-ai","claimed":false,"request_supported":{"post":"https://gitbuyer.com/r/lightly-ai/lightly-train/request-supported","requests":0},"note":"indexed from public GitHub; nothing is for sale on this page. Clone it from GitHub. Paid listings live at /search."}