ai/com.siddharth.kmp.ai/InjectableNativeLlm

InjectableNativeLlm

class InjectableNativeLlm : NativeLlm(source)

Generic delegate-or-degrade injection seam, same shape as Doori's InjectableDocumentAiAnalyzer/ InjectableTextGenerator — kept in commonMain (not iosMain) so this logic is unit-testable; the iosMain holder (FoundationModelsBridge) has nothing left to test once it just forwards here.

Constructors

InjectableNativeLlm

constructor()

Properties

generator

var generator: NativeLlm?

Set once at app startup (e.g. from AppDelegate) by the platform bridge — see ai/ios-bridge/README.md.

Functions

generate

open suspend override fun generate(prompt: String): String?

Null means "declined" — unavailable, or a runtime failure — never throws. Matches OnDeviceLlm.generate's no-throw contract, and keeps a Swift implementation from having to propagate a native error across the interop boundary.

generateStream

open override fun generateStream(prompt: String, callback: NativeLlmStreamCallback): NativeLlmCancelHandle

Streams prompt token-by-token through callback — real per-token output (Apple's LanguageModelSession.streamResponse), not a single-emission replay of generate. Returns a NativeLlmCancelHandle the Kotlin side calls when its collecting coroutine is cancelled, so Stop/screen-leave actually halts native generation instead of only stopping the Kotlin side from listening — the same contract MediaPipeOnDeviceLlm's Android backend already honors for its own callbackFlow.

isAvailable

open override fun isAvailable(): Boolean