apple-on-device-ai

On-Device AI for Apple Platforms

Guide for selecting, deploying, and optimizing on-device ML models. Covers Apple Foundation Models, Core ML, MLX Swift, and llama.cpp.

Contents

• Framework Selection Router
• Apple Foundation Models Overview
• Core ML Overview
• MLX Swift Overview
• Multi-Backend Architecture
• Performance Best Practices
• Common Mistakes
• Review Checklist
• References

Framework Selection Router

Use this decision tree to pick the right framework for your use case.

Apple Foundation Models

When to use: Text generation, summarization, entity extraction, structured output, and short dialog on iOS 26+ / macOS 26+ devices with Apple Intelligence enabled. No app-managed API key, network round trip, or model hosting; still handle system model asset readiness.

Best for:

• Generating text or structured data with @Generable types
• Summarization, classification, content tagging
• Tool-augmented generation with the Tool protocol
• Apps that need guaranteed on-device privacy

Not suited for: Complex math, code generation, factual accuracy tasks, or apps targeting pre-iOS 26 devices.

Core ML

When to use: Deploying custom trained models (vision, NLP, audio) across all Apple platforms. Converting models from PyTorch, TensorFlow, or scikit-learn with coremltools.

Best for:

• Image classification, object detection, segmentation
• Custom NLP classifiers, sentiment analysis models
• Audio/speech models via SoundAnalysis integration
• Any scenario needing Neural Engine optimization
• Models requiring quantization, palettization, or pruning

MLX Swift

When to use: Running specific open-source LLMs (Llama, Mistral, Qwen, Gemma) on Apple Silicon with maximum throughput. Research and prototyping.

Best for:

• Highest sustained token generation on Apple Silicon
• Running Hugging Face models from mlx-community
• Research requiring automatic differentiation
• Fine-tuning workflows on Mac

llama.cpp

When to use: Cross-platform LLM inference using GGUF model format. Production deployments needing broad device support.

Best for:

• GGUF quantized models (Q4_K_M, Q5_K_M, Q8_0)
• Cross-platform apps (iOS + Android + desktop)
• Maximum compatibility with open-source model ecosystem

Quick Reference

Apple Foundation Models Overview

Use the system language model for short generation, summarization, tagging, structured output, and tool-augmented tasks on Apple Intelligence devices. Gate every entry point before creating a session:

import FoundationModels
 
switch SystemLanguageModel.default.availability {
case .available:
    guard SystemLanguageModel.default.

Then create a session and keep its shared context budget small:

Required guardrails:

• Sessions are stateful and accept one request at a time; serialize access and check isResponding before issuing another response.
• Instructions, tools, schemas, prompts, transcripts, and output share the context window. Register only necessary tools and keep schemas compact.
• Resolve the locale with supportsLocale(_:); do not raw-match language lists.
• Keep untrusted user content in prompts, never instructions. System guardrails remain active, so handle refusal and other generation errors with fallback UI.

Load the Foundation Models reference when the task needs @Generable, @Guide, streaming, tool definitions, transcripts, generation options, custom adapters, prompt design, or detailed error handling.

Core ML Overview

Apple's framework for deploying trained models. Automatically dispatches to the optimal compute unit (CPU, GPU, or Neural Engine).

Model Formats

Always use mlprogram (.mlpackage) for new work.

Conversion Pipeline (coremltools)

Validate, Fix, and Reconvert

1. Freeze representative source-model fixtures and acceptable output/task tolerances before conversion.
2. Convert, then run the same fixtures through the source and Core ML models.
3. If output parity or task metrics miss tolerance, inspect shapes, operators, precision, and preprocessing; fix the conversion and rerun the fixtures.
4. Compress only after the uncompressed model passes. Revalidate accuracy after each compression change and undo or tune changes that miss the threshold.
5. Profile the passing model on physical target devices, then repeat until correctness, latency, memory, and package-size targets all pass.

Boundary with coreml

This skill owns Python-side conversion, compression, profiling, and framework selection. Use the sibling coreml skill for Swift app integration, prediction APIs, runtime configuration, Vision request wiring, and detailed model loading.

See references/coreml-conversion.md for the full conversion pipeline and references/coreml-optimization.md for optimization techniques.

MLX Swift Overview

Apple's ML framework for Swift. Highest sustained generation throughput on Apple Silicon via unified memory architecture.

Loading and Running LLMs

Model Selection by Device

Memory Management

1. Never exceed 60% of total RAM on iOS
2. Set MLX cache limits: Memory.cacheLimit = 512 * 1024 * 1024
3. Unload MLX and llama.cpp models on backgrounding or memory pressure; for MLX, also call Memory.clearCache() after generation-heavy phases
4. Use "Increased Memory Limit" entitlement for larger models
5. Validate MLX Swift and llama.cpp on physical Apple Silicon; Simulator cannot exercise Metal-dependent inference, memory, or performance

See references/mlx-swift.md for full MLX Swift patterns and llama.cpp integration.

Multi-Backend Architecture

When an app needs multiple AI backends (e.g., Foundation Models + MLX fallback):

Serialize all model access through a coordinator actor to prevent contention:

For custom Core ML models, name only the conversion/optimization handoff here: send Swift app integration, model loading, Vision wiring, and prediction lifecycle to coreml. Keep private user content, such as journals, on device unless product explicitly opts into a nonlocal fallback.

Performance Best Practices

1. Run outside debugger for accurate benchmarks (Xcode: Cmd-Opt-R, uncheck "Debug Executable")
2. Call session.prewarm() for Foundation Models before user interaction
3. Pre-compile Core ML models to .mlmodelc for faster loading
4. Use EnumeratedShapes over RangeDim for Neural Engine optimization
5. Use 4-bit palettization for best Neural Engine memory/latency gains
6. Hand off detailed Vision, Natural Language, and Swift Core ML runtime integration to the sibling framework skills

Common Mistakes

1. No availability check. Starting generation without checking SystemLanguageModel.default.availability leaves unsupported devices with failures instead of fallback UI.
2. No fallback UI. Users on pre-iOS 26 or devices without Apple Intelligence see nothing. Always provide a graceful degradation path.
3. Exceeding the context window. The token budget covers input + output. Monitor usage via tokenCount(for:) and summarize when needed.
4. Concurrent requests on one session. LanguageModelSession supports one request at a time. Check session.isResponding or serialize access.
5. Untrusted content in instructions. User input placed in the instructions parameter bypasses guardrail boundaries. Keep user content in the prompt.
6. Skipping conversion parity checks. Compare the source and Core ML model on fixed fixtures, then fix and reconvert before compressing or shipping.
7. Forgetting model.eval() before Core ML tracing. PyTorch models must be in eval mode before torch.jit.trace. Training-mode artifacts corrupt output.
8. Using neuralnetwork format. Always use mlprogram (.mlpackage) for new Core ML models. The legacy neuralnetwork format is deprecated.
9. Exceeding 60% RAM on iOS (MLX Swift). Large models cause OOM kills.
10. Trusting MLX simulator results. Validate Metal-dependent behavior on physical devices; Simulator is only a UI/control-flow smoke test.
11. Not clearing MLX caches. Pair model unload with Memory.clearCache().

Review Checklist

• Framework selection matches use case and target OS version
• Foundation Models: availability checked before every API call
• Foundation Models: graceful fallback when model unavailable
• Foundation Models: session prewarm called before user interaction
• Foundation Models: @Generable properties in logical generation order
• Foundation Models: token budget accounted for (check contextSize)
• Core ML: model format is mlprogram (.mlpackage) for iOS 15+
• Core ML: source/Core ML parity passes fixed fixtures and task tolerances
• Core ML: compressed model revalidated and profiled on physical targets
• MLX Swift: model size appropriate for target device RAM
• MLX Swift: cache limits set, caches cleared, models unloaded
• All model access serialized through coordinator actor
• Concurrency: model types and tool implementations are Sendable-conformant or @MainActor-isolated
• Physical device testing performed (not simulator)

References

• Foundation Models API -- LanguageModelSession, @Generable, tool calling, prompt design
• Core ML Conversion -- Model conversion from PyTorch, TensorFlow, other frameworks
• Core ML Optimization -- Quantization, palettization, pruning, performance tuning
• MLX Swift & llama.cpp -- MLX Swift patterns, llama.cpp integration, memory management