Instructions to use TheCluster/YOLOv11-CoreML with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- ultralytics
How to use TheCluster/YOLOv11-CoreML with ultralytics:
from huggingface_hub import hf_hub_download from ultralytics import YOLO # pick the weights file from this repo's "Files and versions" tab weights = hf_hub_download("TheCluster/YOLOv11-CoreML", "<weights>.pt") model = YOLO(weights) source = 'http://images.cocodataset.org/val2017/000000039769.jpg' model.predict(source=source, save=True) - Notebooks
- Google Colab
- Kaggle
File size: 462 Bytes
d9c70cb | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 | ---
license: agpl-3.0
pipeline_tag: object-detection
tags:
- yolo
- yolov11
- apple
- coreml
- ane
- tracking
- ultralytics
---
### YOLOv11 CoreML
YOLOv11 models converted to CoreML format.
Using:
- pytorch 2.5.0
- coremltools 8.0
- ultralytics 8.3.16
- macOS Sequoia 15.1
Tested on:
- M1
- M1 Ultra
- M4 (iPad Pro 2024)
- A15 (iPhone 14)
Models can be run using Apple Neural Engine
Model sources: https://github.com/ultralytics/ultralytics |