AutoYou
Freemium
Users wanting fully private local AI
Core AI Models
FreeOpen SourceSelf-Hosted
Getting a PyTorch model running on an iPhone used to mean guessing at export flags, fighting Core ML type mismatches, and hoping the runtime behaved the same way the Python prototype did — Apple's coreai-models exists to close that gap.
Similar tools
llama.cppFreellama.cpp is the MIT-licensed C/C++ engine that runs GGUF language models on your own hardware…
LocalAIFreeLocalAI is an MIT-licensed server that exposes an OpenAI-compatible REST API from your own…
AutoYouFreemiumAutoYou runs a local AI server on your Windows, macOS, or Linux machine, then lets your phone…
LM-Kit OneFreemiumThe vendor describes LM-Kit One as a self-hosted AI application server that exposes OpenAI…
Get This Tool
License: BSD-3-Clause Any use incl. commercial
Local-run terms: Clone repository, use uv to run export recipes and Swift package on macOS/iOS with Xcode; models run via Core AI framework.
Pricing
- Model
- Free
Summary
The repository ships three concrete layers: Python export recipes for popular Hugging Face models, reusable PyTorch primitives for authoring custom models in Core AI format, and a Swift package that slots those exported models into macOS and iOS apps. The CLI tooling lets you run models directly on a Mac before touching Xcode. Where the workflow breaks is at the edges of what the export recipes cover — models outside the supported Hugging Face roster require you to author your own export logic using the Python primitives, which assumes familiarity with both PyTorch internals and Core AI's model format. The skills directory adds coding-agent plugins, but the core offering is an export-and-runtime pipeline, not an autonomous agent loop.
Bottom line: Reach for this when your target is Apple silicon and your model is on the supported export list; plan for manual primitive authoring the moment you step outside that list.
Find your fit
Before you choose Core AI Models
More options in Local Inference Runtimes. Compare pricing, platforms, and features side by side.
LM-Kit One
Freemium
Companies needing data-residency compliance
Pros
- Export recipes for popular Hugging Face models are included out of the box, so you skip the format-guessing phase that typically consumes the first day of any on-device ML project.
- The Swift runtime package is built directly on Core AI framework and lives in the same repo as the export tooling, which means the Python-to-Swift handoff follows a maintained path rather than an improvised one.
- Reusable PyTorch primitives for custom model authoring give you a structured starting point when your architecture is not covered by the existing recipes, rather than a blank canvas.
- CLI tooling for local Mac inference lets you validate model behavior before opening Xcode, catching export problems before they become app-integration problems.
- BSD-3-Clause license and a fully public GitHub repository mean you can fork, audit, and modify the export logic — critical when Apple silicon deployment has compliance or reproducibility requirements.
Cons
- Models outside the supported Hugging Face export recipes require writing custom export logic with the Python primitives; this is not a guided path, and teams without PyTorch internals experience stall here and move to ONNX-based pipelines with broader model coverage.
- There is no API and no hosted runtime — everything runs from a locally cloned repository, so teams expecting a managed service or cloud-side inference endpoint abandon this and use a hosted inference provider instead.
- The tool produces Core AI format artifacts, which are not portable outside the Apple ecosystem; any project that also targets Android or web inference requires a parallel export pipeline, meaning two separate toolchains to maintain.
About
- Platforms
- macOS, iOS
- API Available
- No
- Self-Hosted
- Yes
- Last Updated
- June 18, 2026
Best For
Who it's for
- Developers targeting Apple silicon on-device AI
- Users needing PyTorch model export to proprietary format
- Swift-based app integration of AI models
What it does well
- Exporting Hugging Face models to Core AI format
- Authoring custom PyTorch models for on-device execution
- Integrating exported models into macOS and iOS apps via Swift
- Running models directly on Mac with included CLI tools
Integrations
Frequently Asked Questions
- Is Core AI Models free?
- Core AI Models offers a free option. Hosting, model usage, or optional services may have separate costs; check the pricing details and official terms.
- Is Core AI Models open source?
- Yes. Core AI Models is open source.
- Can I self-host Core AI Models?
- Yes. Core AI Models supports self-hosting on your own infrastructure.
- What platforms does Core AI Models support?
- Core AI Models is available on: macOS, iOS.
Curated lists that include this category
coreai-models is Apple’s open-source toolkit — BSD-3-Clause licensed, hosted publicly on GitHub — for converting models into the Core AI format required for on-device execution on Apple silicon. The pipeline moves in three steps: export a model from Hugging Face or another source using the included Python recipes, optionally author custom model logic using the reusable PyTorch primitives, then consume the exported artifact through the Swift package in a macOS or iOS application. A CLI layer lets developers test model inference locally on a Mac before any Xcode work begins.
The differentiating feature is the combination of export recipes and a first-party Swift runtime package in a single repository. Most third-party export tools stop at the file conversion step and leave the app-side integration to you. Here the Swift package is built directly on top of the Core AI framework, which means the handoff from Python export to Swift consumption follows a defined path the vendor maintains — rather than one you improvise per project.
The tool fits tightly scoped scenarios: a developer targeting Apple silicon who needs a supported Hugging Face model running in a native Swift app. It breaks when the model architecture falls outside the recipe library. At that point, teams write their own export logic using the Python primitives, which is documented but assumes non-trivial familiarity with PyTorch model internals. There is no API, no hosted service, and no GUI — the entire workflow runs locally from a cloned repository using uv-based Python tooling and Swift Package Manager.
The repository uses uv for Python dependency management and Swift Package Manager for the runtime side, with pre-commit hooks and a .swift-format config signaling active maintenance standards. The skills directory contains plugins designed to help coding agents work with Core AI effectively, though the vendor describes these as secondary to the export-and-runtime core.
Contribute
- Work on Core AI Models? Claim this listing.
- Embed a badge for your README or site.
Sign in to rate this tool or suggest an edit.