ML Models for iOS

A curated collection of models optimized for on-device execution using Core ML. These models are ready to be integrated into your Swift projects.

Model 🔼	Type ↕️	Source	Format	Size ↕️	Action
BERT-SQuAD Find answers to questions about paragraphs of text using the SQuAD dataset.	Natural Language	Apple	ML Package	420 MB	Download
DETR Resnet50 DEtection TRansformer (DETR) model for object detection and panoptic segmentation.	Vision	Apple	Core ML	167 MB	Download
DeepLabv3 Segment the pixels of a camera frame or image into a predefined set of classes.	Vision	Apple	Core ML	9.6 MB	Download
Depth Anything V2 The Depth Anything model performs monocular depth estimation.	Vision	Apple	Core ML	Varies	Download
FastViT A Fast Hybrid Vision Transformer architecture trained to classify the dominant object in a camera frame or image.	Vision	Apple	Core ML	Varies	Download
MNIST Classify a single handwritten digit (supports digits 0-9).	Vision	Apple	Core ML	4.8 MB	Download
MobileNetV2 Efficient architecture for image classification, optimized for mobile devices.	Vision	Apple	Core ML	13.1 MB	Download
ResNet-50 A Residual Neural Network that classifies the dominant object in an image.	Vision	Apple	Core ML	102.6 MB	Download
Updatable Drawing Classifier Recognize new drawings based on a K-Nearest Neighbors model (KNN).	Vision	Apple	Core ML	Varies	Download
YOLOv3 Locate and classify 80 different types of objects in real-time.	Vision	Apple	Core ML	248.4 MB	Download