…ing ConvNeXt V2, RT-DETRv2, EoMT, Mask2Former and DINOv3-based variants of several models. - Ultralytics YOLO models – are now available for everyone in pre-built Docker

…- Add opt-in activation checkpointing for DINOv2, DINOv3, and ECViT backbones during pretraining with activation checkpoint args={"enabled": True} . Reduces peak GPU memo

…a2 option runs in-process through ONNX Runtime: a DINOv3 descriptor graph with on-device retrieval pooling and a bidirectional coarse-match graph, preset base (640 px), f

…ce invalid teacher features. Removed - Remove the DINOv3.1 pretraining method. Fixed - Preserve exponential-moving-average updates when resuming training, preventing a sp

…s://github.com/Robbyant/lingbot-vision) backbones dinov3/vits16-lingbot , dinov3/vitb16-lingbot , and dinov3/vitl16-lingbot . - Add LingBot Vision backbones to the DINOv3

…e Kwon](https://github.com/developer0hye) Support DINOv3 weights in NaFlexVit. Thanks [Yonghye Kwon](https://github.com/developer0hye) Some improvements to Muon fallback