Pre-trained Vision ModelSAM 2by Meta AITasksegmentationArchitectureTransformer with streaming memory (promptable, video-capable)FrameworkPyTorchSign in to save SAM 2