# apple/coreai-models

Model export recipes, Python primitives, and Swift runtime utilities for on-device AI

Repository: https://github.com/apple/coreai-models
Canonical: https://ross.abutalabs.com/products/coreai-models
Language: Swift
License: BSD-3-Clause
License Family: permissive
Last push: 2026-08-26T00:44:59+00:00

## Health v2 (maintenance only)
Score: 67/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 60, longevity 6
- inputs: {"age_days": 86, "days_push": 8, "days_rel": 56, "gap_med": null, "n_releases_24m": 1}
- flags: prerelease_only, young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1999, forks 183 (observed 2026-08-28T04:06:03.523259+00:00)

## What it is
Apple's repository of model export recipes, Python primitives, and Swift runtime utilities for building on-device AI with the Core AI framework. It converts popular open-source models (e.g., from Hugging Face) into standalone .aimodel files that run on macOS and iOS via Apple silicon.

## Use cases
- export huggingface models to run on-device on iphone
- convert pytorch models to core ai format
- run llms locally on a mac
- integrate on-device ai models into an ios app
- author custom models for apple core ai in pytorch
- find supported models for core ai export
- help coding agents use core ai effectively

## When to choose
- you want to deploy open-source models on-device on macOS or iOS with Apple's Core AI framework
- you need export recipes for popular Hugging Face models to .aimodel format
- you're building an Apple-silicon app that runs language or diffusion models locally
- you want Swift runtime utilities for model integration in Xcode projects

## When to avoid
- you target Android, Windows, or Linux deployment
- you need server-side or cloud GPU inference
- you can't upgrade to macOS/iOS 27 and Xcode 27
- you need a framework-agnostic inference solution independent of Apple's ecosystem

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, llm-inference, sdk, cli, developer-tools
- domain: machine-learning, artificial-intelligence, large-language-models, apple-ecosystem, developer-tools
- platform: python, cli
- tags: on-device-ai, core-ai, model-export, apple-silicon, pytorch, aimodel, swift-package, agent-skills, macos, ios, swift

## Member repositories
- apple/coreai-models (main) score 67

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:03.523259+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:02:10.034761+00:00, confidence not recorded.
  - readme: https://github.com/apple/coreai-models (fetched 2026-08-28T04:06:03.523259+00:00, sha 2e76de9151a0)
- Data as of 2026-08-30T08:39:29.467469+00:00.