apple-foundation-modelslisted

# Apple FoundationModels (on-device LLM, iOS 26+) The FoundationModels framework runs Apple Intelligence's language model entirely on-device: no network, no data leaving the device. Key constraints: availability depends on device eligibility and user settings, and the context window is 4,096 tokens total (instructions + prompt + output combined) — chunk larger inputs across sessions. ## Availability Check `SystemLanguageModel.default.availability` before creating a session; every unavailability case is reachable in the field: ```swift struct GenerativeView: View { private var model = SystemLanguageModel.default var body: some View { switch model.availability { case .available: ContentView() case .unavailable(.deviceNotEligible): Text("Device not eligible for Apple Intelligence") case .unavailable(.appleIntelligenceNotEnabled): Text("Please enable Apple Intelligence in Settings") case .unavailable(.modelNotReady): Text("Model is downloading or not ready") case .unavailable(let other): Text("Model unavailable: \(other)") } } } ``` ## LanguageModelSession ```swift // Single-turn: new session each time let session = LanguageModelSession() let response = try await session.respond(to: "What's a good month to visit Paris?") print(response.content) // Multi-turn: reuse the session; it keeps conversation context let session = LanguageModelSession(inst