Axiom Foundation Models Ref

Foundation Models Framework — Complete API Reference

Overview

The Foundation Models framework provides access to Apple's on-device Large Language Model (3 billion parameters, 2-bit quantized) with a Swift API. This reference covers every API, all WWDC 2025 code examples, and comprehensive implementation patterns.

Model Specifications

Technical Details:

• Parameters: 3 billion (3B)
• Quantization: 2-bit
• Context Window: 4096 tokens (combined input + output)
• Supported Languages: English + additional languages via SystemLanguageModel.default.supportedLanguages
• Platforms: iOS 26+, macOS 26+, iPadOS 26+, axiom-visionOS 26+

Optimized For:

• Text summarization
• Information extraction
• Content classification
• Content generation
• Tag generation
• Entity detection

NOT Optimized For:

• World knowledge queries
• Complex multi-step reasoning
• Mathematical computation
• Translation (use dedicated translation models)

Privacy & Performance:

• Runs entirely on-device
• No network required (works offline)
• Data never leaves device
• No per-request costs
• Integrated into OS (doesn't increase app size)

When to Use This Reference

Use this reference when:

• Implementing Foundation Models features
• Understanding API capabilities
• Looking up specific code examples
• Planning architecture with Foundation Models
• Migrating from prototype to production
• Debugging implementation issues

Related Skills:

• axiom-foundation-models — Discipline skill with anti-patterns, pressure scenarios, decision trees
• axiom-foundation-models-diag — Diagnostic skill for troubleshooting issues

LanguageModelSession

Overview

LanguageModelSession is the core class for interacting with the model. It maintains conversation history (transcript), handles multi-turn interactions, and manages model state.

Creating a Session

Basic Creation:

import FoundationModels

let session = LanguageModelSession()

With Custom Instructions:

let session = LanguageModelSession(instructions: """
    You are a friendly barista in a pixel art coffee shop.
    Respond to the player's question concisely.
    """
)

From WWDC 301:1:05

With Tools:

let session = LanguageModelSession(
    tools: [GetWeatherTool()],
    instructions: "Help user with weather forecasts."
)

From WWDC 286:15:03

With Specific Model/Use Case:

let session = LanguageModelSession(
    model: SystemLanguageModel(useCase: .contentTagging)
)

From WWDC 286:18:39

Instructions vs Prompts

Instructions:

• Come from developer
• Define model's role, style, constraints
• Mostly static
• First entry in transcript
• Model trained to obey instructions over prompts (security feature)

Prompts:

• Come from user (or dynamic app state)
• Specific requests for generation
• Dynamic input
• Each call to respond(to:) adds prompt to transcript

Security Consideration:

• NEVER interpolate untrusted user input into instructions
• User input should go in prompts only
• Prevents prompt injection attacks

respond(to:) Method

Basic Text Generation:

func respond(userInput: String) async throws -> String {
    let session = LanguageModelSession(instructions: """
        You are a friendly barista in a world full of pixels.
        Respond to the player's question.
        """
    )
    let response = try await session.respond(to: userInput)
    return response.content
}

From WWDC 301:1:05

Return Type: Response<String> with .content property

respond(to:generating:) Method

Structured Output with @Generable:

@Generable
struct SearchSuggestions {
    @Guide(description: "A list of suggested search terms", .count(4))
    var searchTerms: [String]
}

let prompt = """
    Generate a list of suggested search terms for an app about visiting famous landmarks.
    """

let response = try await session.respond(
    to: prompt,
    generating: SearchSuggestions.self
)

print(response.content) // SearchSuggestions instance

From WWDC 286:5:51

Return Type: Response<SearchSuggestions> with .content property

Generation Options

Deterministic Output (Greedy Sampling):

let response = try await session.respond(
    to: prompt,
    options: GenerationOptions(sampling: .greedy)
)

From WWDC 301:6:14

Low Variance (Conservative):

let response = try await session.respond(
    to: prompt,
    options: GenerationOptions(temperature: 0.5)
)

High Variance (Creative):

let response = try await session.respond(
    to: prompt,
    options: GenerationOptions(temperature: 2.0)
)

From WWDC 301:6:14

Skip Schema in Prompt (Optimization):

let response = try await session.respond(
    to: prompt,
    generating: Person.self,
    options: GenerationOptions(includeSchemaInPrompt: false)
)

Use when: Subsequent requests with same @Generable type. Reduces token count and latency.

Multi-Turn Interactions

Retaining Context

let session = LanguageModelSession()

// First turn
let firstHaiku = try await session.respond(to: "Write a haiku about fishing")
print(firstHaiku.content)
// Silent waters gleam,
// Casting lines in morning mist—
// Hope in every cast.

// Second turn - model remembers context
let secondHaiku = try await session.respond(to: "Do another one about golf")
print(secondHaiku.content)
// Silent morning dew,
// Caddies guide with gentle words—
// Paths of patience tread.

print(session.transcript) // Shows full history

From WWDC 286:17:46

How it works:

• Each respond() call adds entry to transcript
• Model uses entire transcript for context
• Enables conversational interactions

Transcript Property

let transcript = session.transcript

for entry in transcript.entries {
    print("Entry: \(entry.content)")
}

Use cases:

• Debugging generation issues
• Displaying conversation history in UI
• Exporting chat logs
• Condensing for context management

isResponding Property

struct HaikuView: View {
    @State private var session = LanguageModelSession()
    @State private var haiku: String?

    var body: some View {
        if let haiku {
            Text(haiku)
        }

        Button("Go!") {
            Task {
                haiku = try await session.respond(
                    to: "Write a haiku about something you haven't yet"
                ).content
            }
        }
        // Gate on `isResponding`
        .disabled(session.isResponding)
    }
}

From WWDC 286:18:22

Why important: Prevents multiple concurrent requests, which could cause errors or unexpected behavior.

@Generable Macro

Overview

@Generable enables structured output from the model using Swift types. The macro generates a schema at compile-time and uses constrained decoding to guarantee structural correctness.

Basic Usage

On Structs:

@Generable
struct Person {
    let name: String
    let age: Int
}

let response = try await session.respond(
    to: "Generate a person",
    generating: Person.self
)

let person = response.content // Type-safe Person instance

From WWDC 301:8:14

On Enums:

@Generable
struct NPC {
    let name: String
    let encounter: Encounter

    @Generable
    enum Encounter {
        case orderCoffee(String)
        case wantToTalkToManager(complaint: String)
    }
}

From WWDC 301:10:49

Supported Types

Primitives:

• String
• Int, Float, Double, Decimal
• Bool

Collections:

• [ElementType] (arrays)

Composed Types:

@Generable
struct Itinerary {
    var destination: String
    var days: Int
    var budget: Float
    var rating: Double
    var requiresVisa: Bool
    var activities: [String]
    var emergencyContact: Person
    var relatedItineraries: [Itinerary] // Recursive!
}

From WWDC 286:6:18

@Guide Constraints

Natural Language Description:

@Generable
struct NPC {
    @Guide(description: "A full name")
    let name: String
}

From WWDC 301:11:20

Numeric Range:

@Generable
struct Character {
    @Guide(.range(1...10))
    let level: Int
}

From WWDC 301:11:20

Array Count:

@Generable
struct Suggestions {
    @Guide(.count(3))
    let attributes: [Attribute]
}

From WWDC 301:11:20

Array Maximum Count:

@Generable
struct Result {
    @Guide(.maximumCount(3))
    let topics: [String]
}

Regex Patterns:

@Generable
struct NPC {
    @Guide(Regex {
        Capture {
            ChoiceOf {
                "Mr"
                "Mrs"
            }
        }
        ". "
        OneOrMore(.word)
    })
    let name: String
}

session.respond(to: "Generate a fun NPC", generating: NPC.self)
// > {name: "Mrs. Brewster"}

From WWDC 301:13:40

Constrained Decoding

How it works:

1. @Generable macro generates schema at compile-time
2. Schema defines valid token sequences
3. During generation, model creates probability distribution for next token
4. Framework masks out invalid tokens based on schema
5. Model can only pick tokens valid according to schema
6. Guarantees structural correctness - no hallucinated keys, no invalid JSON

From WWDC 286: "Constrained decoding prevents structural mistakes. Model is prevented from generating invalid field names or wrong types."

Benefits:

• Zero parsing code needed
• No runtime parsing errors
• Type-safe Swift objects
• Compile-time safety (changes to struct caught by compiler)

Property Declaration Order

Properties generated in order declared:

@Generable
struct Itinerary {
    var name: String        // Generated FIRST
    var days: [DayPlan]     // Generated SECOND
    var summary: String     // Generated LAST
}

Why it matters:

• Later properties can reference earlier ones
• Better model quality: Summaries after content
• Better streaming UX: Important properties first

From WWDC 286:11:00

Streaming

Overview

Foundation Models uses snapshot streaming (not delta streaming). Instead of raw deltas, the framework streams PartiallyGenerated types with optional properties that fill in progressively.

PartiallyGenerated Type

The @Generable macro automatically creates a PartiallyGenerated nested type:

@Generable
struct Itinerary {
    var name: String
    var days: [DayPlan]
}

// Compiler generates:
extension Itinerary {
    struct PartiallyGenerated {
        var name: String?        // All properties optional!
        var days: [DayPlan]?
    }
}

From WWDC 286:9:20

streamResponse Method

@Generable
struct Itinerary {
    var name: String
    var days: [Day]
}

let stream = session.streamResponse(
    to: "Craft a 3-day itinerary to Mt. Fuji.",
    generating: Itinerary.self
)

for try await partial in stream {
    print(partial) // Incrementally updated Itinerary.PartiallyGenerated
}

From WWDC 286:9:40

Return Type: AsyncSequence<Itinerary.PartiallyGenerated>

SwiftUI Integration

struct ItineraryView: View {
    let session: LanguageModelSession
    let dayCount: Int
    let landmarkName: String

    @State
    private var itinerary: Itinerary.PartiallyGenerated?

    var body: some View {
        VStack {
            if let name = itinerary?.name {
                Text(name).font(.title)
            }

            if let days = itinerary?.days {
                ForEach(days, id: \.self) { day in
                    DayView(day: day)
                }
            }

            Button("Start") {
                Task {
                    do {
                        let prompt = """
                            Generate a \(dayCount) itinerary \
                            to \(landmarkName).
                            """

                        let stream = session.streamResponse(
                            to: prompt,
                            generating: Itinerary.self
                        )

                        for try await partial in stream {
                            self.itinerary = partial
                        }
                    } catch {
                        print(error)
                    }
                }
            }
        }
    }
}

From WWDC 286:10:05

Best Practices

1. Use SwiftUI animations:

if let name = itinerary?.name {
    Text(name)
        .transition(.opacity)
}

2. View identity for arrays:

// ✅ GOOD - Stable identity
ForEach(days, id: \.id) { day in
    DayView(day: day)
}

// ❌ BAD - Identity changes
ForEach(days.indices, id: \.self) { index in
    DayView(day: days[index])
}

3. Property order optimization:

// ✅ GOOD - Title first for streaming
@Generable
struct Article {
    var title: String      // Shows immediately
    var summary: String    // Shows second
    var fullText: String   // Shows last
}

From WWDC 286:11:00

Tool Protocol

Overview

Tools let the model autonomously execute your custom code to fetch external data or perform actions. Tools integrate with MapKit, WeatherKit, Contacts, EventKit, or any custom API.

Protocol Definition

protocol Tool {
    var name: String { get }
    var description: String { get }

    associatedtype Arguments: Generable

    func call(arguments: Arguments) async throws -> ToolOutput
}

Example: GetWeatherTool

import FoundationModels
import WeatherKit
import CoreLocation

struct GetWeatherTool: Tool {
    let name = "getWeather"
    let description = "Retrieve the latest weather information for a city"

    @Generable
    struct Arguments {
        @Guide(description: "The city to fetch the weather for")
        var city: String
    }

    func call(arguments: Arguments) async throws -> ToolOutput {
        let places = try await CLGeocoder().geocodeAddressString(arguments.city)
        let weather = try await WeatherService.shared.weather(for: places.first!.location!)
        let temperature = weather.currentWeather.temperature.value

        let content = GeneratedContent(properties: ["temperature": temperature])
        let output = ToolOutput(content)

        // Or if your tool's output is natural language:
        // let output = ToolOutput("\(arguments.city)'s temperature is \(temperature) degrees.")

        return output
    }
}

From WWDC 286:13:42

Attaching Tools to Session

let session = LanguageModelSession(
    tools: [GetWeatherTool()],
    instructions: "Help the user with weather forecasts."
)

let response = try await session.respond(
    to: "What is the temperature in Cupertino?"
)

print(response.content)
// It's 71˚F in Cupertino!

From WWDC 286:15:03

How it works:

1. Session initialized with tools
2. User prompt: "What's Tokyo's weather?"
3. Model analyzes prompt, decides weather data needed
4. Model generates tool call: getWeather(city: "Tokyo")
5. Framework calls call() method
6. Your code fetches real data from API
7. Tool output inserted into transcript
8. Model generates final response using tool output

From WWDC 301: "Model autonomously decides when and how often to call tools. Can call multiple tools per request, even in parallel."

Example: FindContactTool

import FoundationModels
import Contacts

struct FindContactTool: Tool {
    let name = "findContact"
    let description = "Finds a contact from a specified age generation."

    @Generable
    struct Arguments {
        let generation: Generation

        @Generable
        enum Generation {
            case babyBoomers
            case genX
            case millennial
            case genZ
        }
    }

    func call(arguments: Arguments) async throws -> ToolOutput {
        let store = CNContactStore()

        let keysToFetch = [CNContactGivenNameKey, CNContactBirthdayKey] as [CNKeyDescriptor]
        let request = CNContactFetchRequest(keysToFetch: keysToFetch)

        var contacts: [CNContact] = []
        try store.enumerateContacts(with: request) { contact, stop in
            if let year = contact.birthday?.year {
                if arguments.generation.yearRange.contains(year) {
                    contacts.append(contact)
                }
            }
        }

        guard let pickedContact = contacts.randomElement() else {
            return ToolOutput("Could not find a contact.")
        }

        return ToolOutput(pickedContact.givenName)
    }
}

From WWDC 301:18:47

Stateful Tools

Tools can maintain state across calls using class instead of struct:

class FindContactTool: Tool {
    let name = "findContact"
    let description = "Finds a contact from a specified age generation."

    var pickedContacts = Set<String>() // State!

    @Generable
    struct Arguments {
        let generation: Generation

        @Generable
        enum Generation {
            case babyBoomers
            case genX
            case millennial
            case genZ
        }
    }

    func call(arguments: Arguments) async throws -> ToolOutput {
        // Fetch contacts...
        contacts.removeAll(where: { pickedContacts.contains($0.givenName) })

        guard let pickedContact = contacts.randomElement() else {
            return ToolOutput("Could not find a contact.")
        }

        pickedContacts.insert(pickedContact.givenName) // Update state
        return ToolOutput(pickedContact.givenName)
    }
}

From WWDC 301:21:55

Why: Tool instance persists for session lifetime. Can track what's been called.

Example: GetContactEventTool

import FoundationModels
import EventKit

struct GetContactEventTool: Tool {
    let name = "getContactEvent"
    let description = "Get an event with a contact."

    let contactName: String

    @Generable
    struct Arguments {
        let day: Int
        let month: Int
        let year: Int
    }

    func call(arguments: Arguments) async throws -> ToolOutput {
        // Fetch events from Calendar...
        let eventStore = EKEventStore()
        // ... implementation ...
        return ToolOutput(/* event details */)
    }
}

From WWDC 301:22:27

ToolOutput

Two forms:

1. Natural language (String):

return ToolOutput("Temperature is 71°F")

2. Structured (GeneratedContent):

let content = GeneratedContent(properties: ["temperature": 71])
return ToolOutput(content)

Tool Naming Best Practices

DO:

• Short, readable names: getWeather, findContact
• Use verbs: get, find, fetch, create
• One sentence descriptions
• Keep descriptions concise (they're in prompt)

DON'T:

• Abbreviations: gtWthr
• Implementation details in description
• Long descriptions (increases token count)

From WWDC 301: "Tool name and description put verbatim in prompt. Longer strings mean more tokens, which increases latency."

Multiple Tools

let session = LanguageModelSession(
    tools: [
        GetWeatherTool(),
        FindRestaurantTool(),
        FindHotelTool()
    ],
    instructions: "Plan travel itineraries."
)

// Model autonomously decides which tools to call and when

Tool Calling Behavior

Key facts:

• Tool can be called multiple times per request
• Multiple tools can be called in parallel
• Model decides when to call (not guaranteed to call)
• Arguments guaranteed valid via @Generable

From WWDC 301: "When tools called in parallel, your call method may execute concurrently. Keep this in mind when accessing data."

Dynamic Schemas

Overview

DynamicGenerationSchema enables creating schemas at runtime instead of compile-time. Useful for user-defined structures, level creators, or dynamic forms.

Creating Dynamic Schemas

@Generable
struct Riddle {
    let question: String
    let answers: [Answer]

    @Generable
    struct Answer {
        let text: String
        let isCorrect: Bool
    }
}

From WWDC 301:14:50

If this structure is only known at runtime:

struct LevelObjectCreator {
    var properties: [DynamicGenerationSchema.Property] = []

    mutating func addStringProperty(name: String) {
        let property = DynamicGenerationSchema.Property(
            name: name,
            schema: DynamicGenerationSchema(type: String.self)
        )
        properties.append(property)
    }

    mutating func addBoolProperty(name: String) {
        let property = DynamicGenerationSchema.Property(
            name: name,
            schema: DynamicGenerationSchema(type: Bool.self)
        )
        properties.append(property)
    }

    mutating func addArrayProperty(name: String, customType: String) {
        let property = DynamicGenerationSchema.Property(
            name: name,
            schema: DynamicGenerationSchema(
                arrayOf: DynamicGenerationSchema(referenceTo: customType)
            )
        )
        properties.append(property)
    }

    var root: DynamicGenerationSchema {
        DynamicGenerationSchema(
            name: name,
            properties: properties
        )
    }
}

// Create riddle schema
var riddleBuilder = LevelObjectCreator(name: "Riddle")
riddleBuilder.addStringProperty(name: "question")
riddleBuilder.addArrayProperty(name: "answers", customType: "Answer")

// Create answer schema
var answerBuilder = LevelObjectCreator(name: "Answer")
answerBuilder.addStringProperty(name: "text")
answerBuilder.addBoolProperty(name: "isCorrect")

let riddleDynamicSchema = riddleBuilder.root
let answerDynamicSchema = answerBuilder.root

From WWDC 301:15:10

Validating and Using Dynamic Schemas

let schema = try GenerationSchema(
    root: riddleDynamicSchema,
    dependencies: [answerDynamicSchema]
)

let session = LanguageModelSession()
let response = try await session.respond(
    to: "Generate a fun riddle about coffee",
    schema: schema
)

let generatedContent = response.content // GeneratedContent
let question = try generatedContent.value(String.self, forProperty: "question")
let answers = try generatedContent.value([GeneratedContent].self, forProperty: "answers")

From WWDC 301:15:10

Dynamic vs Static @Generable

Use @Generable when:

• Structure known at compile-time
• Want type safety
• Want automatic parsing

Use Dynamic Schemas when:

• Structure only known at runtime
• User-defined schemas
• Maximum flexibility

From WWDC 301: "Compile-time @Generable gives type safety. Dynamic schemas give runtime flexibility. Both use same constrained decoding guarantees."

Sampling & Generation Options

Sampling Methods

Random Sampling (Default):

let response = try await session.respond(to: prompt)
// Different output each time

Greedy Sampling (Deterministic):

let response = try await session.respond(
    to: prompt,
    options: GenerationOptions(sampling: .greedy)
)
// Same output for same prompt (given same model version)

From WWDC 301:6:14

Use greedy for:

• Unit tests
• Demos that need repeatability
• When consistency critical

Caveat: Only deterministic for same model version. OS updates may change model, changing output.

Temperature Control

Low variance (focused, conservative):

let response = try await session.respond(
    to: prompt,
    options: GenerationOptions(temperature: 0.5)
)
// Predictable, focused output

High variance (creative, diverse):

let response = try await session.respond(
    to: prompt,
    options: GenerationOptions(temperature: 2.0)
)
// Varied, creative output

From WWDC 301:6:14

Temperature scale:

• 0.1-0.5: Very focused
• 1.0 (default): Balanced
• 1.5-2.0: Very creative

Built-in Use Cases

Content Tagging Adapter

Specialized adapter for:

• Tag generation
• Entity extraction
• Topic detection

@Generable
struct Result {
    let topics: [String]
}

let session = LanguageModelSession(
    model: SystemLanguageModel(useCase: .contentTagging)
)

let response = try await session.respond(
    to: articleText,
    generating: Result.self
)

From WWDC 286:19:19

Custom Use Cases

With custom instructions:

@Generable
struct Top3ActionEmotionResult {
    @Guide(.maximumCount(3))
    let actions: [String]
    @Guide(.maximumCount(3))
    let emotions: [String]
}

let session = LanguageModelSession(
    model: SystemLanguageModel(useCase: .contentTagging),
    instructions: "Tag the 3 most important actions and emotions in the given input text."
)

let response = try await session.respond(
    to: text,
    generating: Top3ActionEmotionResult.self
)

From WWDC 286:19:35

Error Handling

GenerationError Types

exceededContextWindowSize:

do {
    let response = try await session.respond(to: prompt)
} catch LanguageModelSession.GenerationError.exceededContextWindowSize {
    // Context limit (4096 tokens) exceeded
    // Solution: Condense transcript, create new session
}

From WWDC 301:3:37

guardrailViolation:

do {
    let response = try await session.respond(to: userInput)
} catch LanguageModelSession.GenerationError.guardrailViolation {
    // Content policy triggered
    // Solution: Show graceful message, don't generate
}

unsupportedLanguageOrLocale:

do {
    let response = try await session.respond(to: userInput)
} catch LanguageModelSession.GenerationError.unsupportedLanguageOrLocale {
    // Language not supported
    // Solution: Check supported languages, show message
}

From WWDC 301:7:06

Context Window Management

Strategy 1: Fresh Session

var session = LanguageModelSession()

do {
    let response = try await session.respond(to: prompt)
    print(response.content)
} catch LanguageModelSession.GenerationError.exceededContextWindowSize {
    // New session, no history
    session = LanguageModelSession()
}

From WWDC 301:3:37

Strategy 2: Condensed Session

do {
    let response = try await session.respond(to: prompt)
} catch LanguageModelSession.GenerationError.exceededContextWindowSize {
    // New session with some history
    session = newSession(previousSession: session)
}

private func newSession(previousSession: LanguageModelSession) -> LanguageModelSession {
    let allEntries = previousSession.transcript.entries
    var condensedEntries = [Transcript.Entry]()

    if let firstEntry = allEntries.first {
        condensedEntries.append(firstEntry) // Instructions

        if allEntries.count > 1, let lastEntry = allEntries.last {
            condensedEntries.append(lastEntry) // Recent context
        }
    }

    let condensedTranscript = Transcript(entries: condensedEntries)
    // Note: transcript includes instructions
    return LanguageModelSession(transcript: condensedTranscript)
}

From WWDC 301:3:55

Availability

Checking Availability

struct AvailabilityExample: View {
    private let model = SystemLanguageModel.default

    var body: some View {
        switch model.availability {
        case .available:
            Text("Model is available").foregroundStyle(.green)
        case .unavailable(let reason):
            Text("Model is unavailable").foregroundStyle(.red)
            Text("Reason: \(reason)")
        }
    }
}

From WWDC 286:19:56

Supported Languages

let supportedLanguages = SystemLanguageModel.default.supportedLanguages
guard supportedLanguages.contains(Locale.current.language) else {
    // Show message
    return
}

From WWDC 301:7:06

Requirements

Device Requirements:

• Apple Intelligence-enabled device
• iPhone 15 Pro or later
• iPad with M1+ chip
• Mac with Apple silicon

Region Requirements:

• Supported region (check Apple Intelligence availability)

User Requirements:

• User opted in to Apple Intelligence in Settings

Performance & Profiling

Foundation Models Instrument

Access: Instruments app → Foundation Models template

Metrics:

• Initial model load time
• Token counts (input/output)
• Generation time per request
• Latency breakdown
• Optimization opportunities

From WWDC 286: "New Instruments profiling template lets you observe areas of optimization and quantify improvements."

Optimization: Prewarming

Problem: First request takes 1-2s to load model

Solution: Create session before user interaction

class ViewModel: ObservableObject {
    private var session: LanguageModelSession?

    init() {
        // Prewarm on init
        Task {
            self.session = LanguageModelSession(instructions: "...")
        }
    }

    func generate(prompt: String) async throws -> String {
        let response = try await session!.respond(to: prompt)
        return response.content
    }
}

From WWDC 259: "Prewarming session before user interaction reduces initial latency."

Time saved: 1-2 seconds off first generation

Optimization: includeSchemaInPrompt

Problem: Large @Generable schemas increase token count

Solution: Skip schema insertion for subsequent requests

// First request - schema inserted
let first = try await session.respond(
    to: "Generate first person",
    generating: Person.self
)

// Subsequent requests - skip schema
let second = try await session.respond(
    to: "Generate another person",
    generating: Person.self,
    options: GenerationOptions(includeSchemaInPrompt: false)
)

From WWDC 259: "Setting includeSchemaInPrompt to false decreases token count and latency for subsequent requests."

Time saved: 10-20% per request

Optimization: Property Order

// ✅ GOOD - Important properties first
@Generable
struct Article {
    var title: String      // Shows in 0.2s (streaming)
    var summary: String    // Shows in 0.8s
    var fullText: String   // Shows in 2.5s
}

// ❌ BAD - Important properties last
@Generable
struct Article {
    var fullText: String   // User waits 2.5s
    var summary: String
    var title: String
}

UX impact: Perceived latency drops from 2.5s to 0.2s with streaming

Feedback & Analytics

LanguageModelFeedbackAttachment

let feedback = LanguageModelFeedbackAttachment(
    input: [
        // Input tokens/prompts
    ],
    output: [
        // Output tokens/content
    ],
    sentiment: .negative,
    issues: [
        LanguageModelFeedbackAttachment.Issue(
            category: .incorrect,
            explanation: "Model hallucinated facts"
        )
    ],
    desiredOutputExamples: [
        [
            // Example of desired output
        ]
    ]
)

let data = try JSONEncoder().encode(feedback)
// Attach to Feedback Assistant report

From WWDC 286:22:13

Use for:

• Reporting quality issues
• Providing examples of desired behavior
• Helping Apple improve models
• Tracking sentiment

Xcode Playgrounds

Overview

Xcode Playgrounds enable rapid iteration on prompts without rebuilding entire app.

Basic Usage

import FoundationModels
import Playgrounds

#Playground {
    let session = LanguageModelSession()
    let response = try await session.respond(
        to: "What's a good name for a trip to Japan? Respond only with a title"
    )
}

From WWDC 286:2:28

Accessing App Types

import FoundationModels
import Playgrounds

#Playground {
    let session = LanguageModelSession()
    for landmark in ModelData.shared.landmarks {
        let response = try await session.respond(
            to: "What's a good name for a trip to \(landmark.name)? Respond only with a title"
        )
    }
}

From WWDC 286:2:43

Benefit: Can access types defined in your app (like @Generable structs)

API Quick Reference

Classes

Protocols

Macros

Enums

Key Methods

Properties

Migration Strategies

From Server LLMs

When to migrate:

• Privacy concerns (data leaving device)
• Offline requirements
• Cost concerns (per-request fees)
• Use case is summarization/extraction/classification

When NOT to migrate:

• Need world knowledge
• Need complex reasoning
• Need very long context (>4096 tokens)

From Manual JSON Parsing

Before:

let prompt = "Generate person as JSON"
let response = try await session.respond(to: prompt)
let data = response.content.data(using: .utf8)!
let person = try JSONDecoder().decode(Person.self, from: data)

After:

@Generable
struct Person {
    let name: String
    let age: Int
}

let response = try await session.respond(
    to: "Generate a person",
    generating: Person.self
)

Benefits:

• No parsing code
• Guaranteed structure
• Type safety
• No crashes from invalid JSON

Resources

WWDC: 286, 259, 301

Skills: axiom-foundation-models, axiom-foundation-models-diag

Last Updated: 2025-12-03 Version: 1.0.0 Skill Type: Reference Content: All WWDC 2025 code examples included