Skip to content
Last Dictionary
Plane
AI Engineering & MLOpsInference & EfficiencyHardware & Compute
trademark

Core ML

proper nounAlso: CoreML, Core ML Tools, coremltools, .mlmodel, .mlpackage, Foundation Models framework

In plain wordsApple's software framework that lets apps on iPhones, iPads and Macs run AI models directly on the device, using its own chips instead of sending data to the cloud.

  1. Apple's framework for integrating trained machine-learning models into apps on iOS, iPadOS, macOS, watchOS, tvOS and visionOS and running them on the device. Models trained in other frameworks such as PyTorchPyTorch, prop. n.: The most widely used free software toolkit for building and training neural networks, especially in research; programmers write ordinary Python and it handles the calculus and the graphics chips. are converted with the open-source coremltools package into Core ML's format (.mlmodel, or the newer .mlpackage bundle), which supports quantizationquantization, n.: Storing an AI model's numbers with fewer digits, like rounding prices to the nearest dollar, so the model takes less memory and runs faster, usually with only a small loss of accuracy., palettization and pruningpruning, n.: Cutting away the parts of an AI model that contribute least, as a gardener trims branches, so the model becomes smaller or faster while still working nearly as well. to shrink models. At run time Core ML decides which parts of a model run on the CPU, the GPU or the Apple Neural Engine (Apple's neural processing unitNPU, abbr.: A small AI chip, or part of a chip, built into phones, laptops and other devices so they can run AI features quickly without draining the battery or using the cloud.), trading speed against power draw and memory use, and Apple's higher-level frameworks for vision, natural language and sound build on it. Core ML serves app developers deploying fixed models rather than researchers training them. Since 2025 Apple has also offered a Foundation Models framework that gives apps direct access to the on-device language model behind Apple Intelligence.

    We converted the segmentation model to Core ML so the app works in airplane mode.

Origin

Announced at Apple's Worldwide Developers Conference in June 2017 and shipped with iOS 11 that year, alongside the first iPhone chip with a Neural Engine, the A11 Bionic. Later versions added on-device personalization of models and support for transformer-based generative models.

Usage note

Contrast MLXMLX, prop. n.: A free Apple software toolkit for running and training AI models efficiently on Mac computers with Apple's own chips, popular for running language models locally on a laptop., Apple's open-source array framework for research and local model running on Macs. Comparable roles on other platforms are played by LiteRTLiteRT, prop. n.: Google's free software for running AI models on phones, small gadgets and microcontrollers; it was called TensorFlow Lite until 2024 and is used inside many Android and iPhone apps. and ExecuTorchExecuTorch, prop. n.: Meta's free software for running PyTorch AI models directly on phones, wearables and other small devices, without needing an internet connection or a data center..

Related

See also
  • on-device AIon-device AI, n.: AI that runs directly on your own phone, laptop, car or headset rather than on a company's distant servers, so it can work offline, respond instantly and keep your data on the device.
  • MLXMLX, prop. n.: A free Apple software toolkit for running and training AI models efficiently on Mac computers with Apple's own chips, popular for running language models locally on a laptop.
  • LiteRTLiteRT, prop. n.: Google's free software for running AI models on phones, small gadgets and microcontrollers; it was called TensorFlow Lite until 2024 and is used inside many Android and iPhone apps.
  • ExecuTorchExecuTorch, prop. n.: Meta's free software for running PyTorch AI models directly on phones, wearables and other small devices, without needing an internet connection or a data center.
  • NPUNPU, abbr.: A small AI chip, or part of a chip, built into phones, laptops and other devices so they can run AI features quickly without draining the battery or using the cloud.
  • ONNXONNX, prop. n.: A shared file format for trained AI models, so that a model built with one toolkit can be moved to and run by another, much as a PDF opens on any computer.
  • edge AIedge AI, n.: Doing AI's computing on or near the devices that collect the data, such as cameras, cars, factory machines, phones or local servers, instead of sending everything off to distant data centers.

Mentioned in

ExecuTorchExecuTorch, prop. n.: Meta's free software for running PyTorch AI models directly on phones, wearables and other small devices, without needing an internet connection or a data center., LiteRTLiteRT, prop. n.: Google's free software for running AI models on phones, small gadgets and microcontrollers; it was called TensorFlow Lite until 2024 and is used inside many Android and iPhone apps., MLXMLX, prop. n.: A free Apple software toolkit for running and training AI models efficiently on Mac computers with Apple's own chips, popular for running language models locally on a laptop., ONNXONNX, prop. n.: A shared file format for trained AI models, so that a model built with one toolkit can be moved to and run by another, much as a PDF opens on any computer.