// Use this skill when working with Apple's Foundation Models framework for on-device AI and LLM capabilities in iOS/macOS apps
| name | Apple Foundation Models |
| description | Use this skill when working with Apple's Foundation Models framework for on-device AI and LLM capabilities in iOS/macOS apps |
| version | 1.0.0 |
Use this skill when you need help with:
This skill covers iOS 26.0+, iPadOS 26.0+, macOS 26.0+, and visionOS 26.0+.
Use this skill when working with Apple's Foundation Models framework for on-device AI and LLM capabilities in iOS/macOS apps. Covers SystemLanguageModel, LanguageModelSession, guided generation, tool calling, and prompting patterns.
TranscriptTranscript.EntryTranscript.ResponseTranscript.ResponseFormatTranscript.SegmentTranscript.StructuredSegmentFoundationModelsGenerationIDSystemLanguageModelSystemLanguageModel.AdapterSystemLanguageModel.GuardrailsSystemLanguageModel.UseCaseSystemLanguageModel.AvailabilityDynamicGenerationSchemaGenerableGenerationGuideGenerationSchemaGuideLanguageModelFeedbackInstructionsInstructionsBuilderInstructionsRepresentableLanguageModelSessionLanguageModelSession.GenerationErrorLanguageModelSession.ToolCallErrorPromptPromptBuilderPromptRepresentableTranscript.InstructionsTranscript.PromptToolAll models run entirely on-device, ensuring privacy and offline capability.
See the references/ directory for detailed API documentation organized by category:
references/advanced.md - Advancedreferences/api_reference.md - Api Referencereferences/getting_started.md - Getting Startedreferences/guided_generation.md - Guided Generationreferences/localization.md - Localizationreferences/prompting.md - Promptingreferences/tool_calling.md - Tool Callinglet model = SystemLanguageModel(useCase: .general)
let session = LanguageModelSession(model: model)
let response = try await session.respond(to: Prompt("Your question"))
struct Recipe: Generable {
let title: String
let ingredients: [String]
}
let recipe = try await session.respond(
generating: Recipe.self,
prompt: Prompt("Create a pasta recipe")
)
struct WeatherTool: Tool {
func call(arguments: String) async throws -> String {
// Fetch weather data
}
}
let session = LanguageModelSession(
model: model,
tools: [WeatherTool()]
)
For complete API details, see the categorized documentation in the references/ directory.