coreml

Name: coreml
Availability: InStock
Author: jperezdelreal

Load Core ML models and run predictions with Swift interfaces.

Updated Apr 6, 2026

Search Tags:#swift #ios #on-device #machine-learning #compute #coreml #mlmodel

Ready to run

npx skills add https://github.com/jperezdelreal/GymBro --skill coreml-jperezdelreal

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: coreml
Source: https://github.com/jperezdelreal/GymBro/tree/main/.squad/skills/ios/coreml
Command: npx skills add https://github.com/jperezdelreal/GymBro --skill coreml-jperezdelreal

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

iOS developers often struggle to integrate and optimize on-device Core ML models, balancing load times, inference performance, and battery life.

Core Features & Use Cases

Loading models from .mlmodelc or .mlpackage for immediate on-device inference.
Predicting outcomes using auto-generated Swift classes and MLFeatureProvider, with support for MLMultiArray and MLTensor.
Configuring compute units (CPU, GPU, Neural Engine) and chaining multi-model pipelines for complex workflows.
Vision integration options via CoreMLRequest (iOS 18+) and VNCoreMLRequest for automatic preprocessing and dispatch.
Deployment patterns including on-device vs on-demand resource strategies.

Quick Start

Create a Swift integration scaffold that loads a model, configures compute units, and runs a prediction on a sample input.

Frequently Asked Questions about coreml

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I load and run predictions with Core ML models in Swift?▼

Load Core ML models from .mlmodelc or .mlpackage files and run predictions using auto-generated Swift classes or MLFeatureProvider with MLMultiArray and MLTensor support for on-device inference.

Can I configure Core ML compute units to use the Neural Engine instead of the CPU?▼

Yes, Core ML compute units can be configured to run on the CPU, GPU, or Neural Engine. This allows you to optimize on-device inference performance, load times, and battery life for iOS apps.

Does Core ML support chaining multiple models together for complex workflows?▼

Core ML supports chaining multi-model pipelines to handle complex workflows. You can integrate multiple on-device models and configure their compute units to build sequential inference tasks.

What's the best way to integrate Vision framework preprocessing with Core ML in iOS?▼

Use CoreMLRequest on iOS 18+ or VNCoreMLRequest for Vision integration. Both options provide automatic preprocessing and dispatch, streamlining image model inference within iOS apps.

What iOS version is required for Core ML Swift integration with MLFeatureProvider?▼

Core ML Swift integration requires Swift 6.3 and iOS 26 or later, while remaining backward-compatible to iOS 14. This environment is necessary for loading models and running predictions.

How do I deploy Core ML models using on-device versus on-demand resource strategies?▼

Core ML deployment patterns include both on-device and on-demand resource strategies. Choose on-device for immediate inference or on-demand to manage storage while balancing load times and battery life.