expo-ai-kit 0.3.2 → 0.3.4

package/android/src/main/java/expo/modules/aikit/GemmaInferenceClient.kt

@@ -1,14 +1,20 @@
 package expo.modules.aikit
 
+import android.app.ActivityManager
 import android.content.Context
 import com.google.ai.edge.litertlm.Engine
 import com.google.ai.edge.litertlm.EngineConfig
 import com.google.ai.edge.litertlm.Conversation
+import com.google.ai.edge.litertlm.ConversationConfig
 import com.google.ai.edge.litertlm.Backend
+import com.google.ai.edge.litertlm.Message
+import com.google.ai.edge.litertlm.MessageCallback
+import com.google.ai.edge.litertlm.Contents
 import kotlinx.coroutines.Dispatchers
 import kotlinx.coroutines.sync.Mutex
 import kotlinx.coroutines.sync.withLock
 import kotlinx.coroutines.withContext
+import kotlinx.coroutines.suspendCancellableCoroutine
 import java.io.File
 import java.io.FileInputStream
 import java.io.FileOutputStream
@@ -16,6 +22,8 @@ import java.io.IOException
 import java.net.HttpURLConnection
 import java.net.URL
 import java.security.MessageDigest
+import kotlin.coroutines.resume
+import kotlin.coroutines.resumeWithException
 
 /**
  * Wrapper around LiteRT-LM Engine for Gemma 4 models.
@@ -46,7 +54,7 @@ class GemmaInferenceClient(private val context: Context) {
    * Unloads any previously loaded model first.
    * Caller is responsible for emitting onModelStateChange events.
    */
-  suspend fun loadModel(modelId: String, modelPath: String) = mutex.withLock {
+  suspend fun loadModel(modelId: String, modelPath: String, minRamBytes: Long = 0, backend: String = "auto") = mutex.withLock {
     // Unload previous model if different
     if (loadedModelId != null && loadedModelId != modelId) {
       conversation?.close()
@@ -60,15 +68,56 @@ class GemmaInferenceClient(private val context: Context) {
       return@withLock // Already loaded
     }
 
+    // Soft memory check. LiteRT-LM memory-maps model weights so actual RSS
+    // is much lower than file size. We log a warning but always attempt the
+    // load — Engine.initialize() may still succeed. If it truly OOMs, Android's
+    // lmkd kills the process (uncatchable signal 9), but that's better than
+    // blocking devices that could have worked.
+    val activityManager = context.getSystemService(Context.ACTIVITY_SERVICE) as? ActivityManager
+    if (activityManager != null && minRamBytes > 0) {
+      val memInfo = ActivityManager.MemoryInfo()
+      activityManager.getMemoryInfo(memInfo)
+      if (memInfo.availMem < minRamBytes) {
+        android.util.Log.w("ExpoAiKit",
+          "Low memory warning for $modelId: " +
+          "available ${memInfo.availMem / 1_000_000}MB, recommended ${minRamBytes / 1_000_000}MB. " +
+          "Attempting load anyway (LiteRT-LM uses memory-mapped I/O)."
+        )
+      }
+    }
+
     try {
       withContext(Dispatchers.IO) {
-        val engineConfig = EngineConfig(
-          modelPath = modelPath,
-          backend = Backend.GPU()
-        )
-        val newEngine = Engine(engineConfig)
-        newEngine.initialize()
-        val newConversation = newEngine.createConversation()
+        val newEngine = when (backend) {
+          "gpu" -> {
+            val config = EngineConfig(modelPath = modelPath, backend = Backend.GPU())
+            val eng = Engine(config)
+            eng.initialize()
+            eng
+          }
+          "cpu" -> {
+            val config = EngineConfig(modelPath = modelPath, backend = Backend.CPU())
+            val eng = Engine(config)
+            eng.initialize()
+            eng
+          }
+          else -> {
+            // "auto": try GPU first, fall back to CPU
+            try {
+              val gpuConfig = EngineConfig(modelPath = modelPath, backend = Backend.GPU())
+              val eng = Engine(gpuConfig)
+              eng.initialize()
+              eng
+            } catch (e: Exception) {
+              val cpuConfig = EngineConfig(modelPath = modelPath, backend = Backend.CPU())
+              val eng = Engine(cpuConfig)
+              eng.initialize()
+              eng
+            }
+          }
+        }
+
+        val newConversation = newEngine.createConversation(ConversationConfig())
 
         engine = newEngine
         conversation = newConversation
@@ -122,7 +171,25 @@ class GemmaInferenceClient(private val context: Context) {
 
     try {
       withContext(Dispatchers.IO) {
-        conv.sendMessage(contents = fullPrompt).toString()
+        suspendCancellableCoroutine<String> { continuation ->
+          val result = StringBuilder()
+          conv.sendMessageAsync(
+            Contents.of(fullPrompt),
+            object : MessageCallback {
+              override fun onMessage(message: Message) {
+                result.clear()
+                result.append(message.toString())
+              }
+              override fun onDone() {
+                continuation.resume(result.toString())
+              }
+              override fun onError(throwable: Throwable) {
+                continuation.resumeWithException(throwable)
+              }
+            },
+            emptyMap()
+          )
+        }
       }
     } catch (e: OutOfMemoryError) {
       throw RuntimeException("INFERENCE_OOM:${loadedModelId ?: "unknown"}:Out of memory during inference")
@@ -135,7 +202,7 @@ class GemmaInferenceClient(private val context: Context) {
    * Generate a streaming response. The onChunk callback receives
    * (token=delta, accumulatedText=full, isDone) matching the PromptApiClient contract.
    *
-   * LiteRT-LM's sendMessageAsync() returns a Flow<Message>. Each emission
+   * LiteRT-LM 0.10.0 uses a MessageCallback interface. Each onMessage emission
    * contains accumulated text, so we diff against previousText to extract
    * the delta token.
    */
@@ -151,21 +218,32 @@ class GemmaInferenceClient(private val context: Context) {
 
     try {
       withContext(Dispatchers.IO) {
-        var previousText = ""
-
-        conv.sendMessageAsync(contents = fullPrompt).collect { message ->
-          val accumulated = message.toString()
-          val token = if (accumulated.length > previousText.length) {
-            accumulated.substring(previousText.length)
-          } else {
-            ""
-          }
-          previousText = accumulated
-          onChunk(token, accumulated, false)
+        suspendCancellableCoroutine<Unit> { continuation ->
+          var previousText = ""
+          conv.sendMessageAsync(
+            Contents.of(fullPrompt),
+            object : MessageCallback {
+              override fun onMessage(message: Message) {
+                val accumulated = message.toString()
+                val token = if (accumulated.length > previousText.length) {
+                  accumulated.substring(previousText.length)
+                } else {
+                  ""
+                }
+                previousText = accumulated
+                onChunk(token, accumulated, false)
+              }
+              override fun onDone() {
+                onChunk("", previousText, true)
+                continuation.resume(Unit)
+              }
+              override fun onError(throwable: Throwable) {
+                continuation.resumeWithException(throwable)
+              }
+            },
+            emptyMap()
+          )
         }
-
-        // Final done event for consistency with PromptApiClient
-        onChunk("", previousText, true)
       }
     } catch (e: OutOfMemoryError) {
       throw RuntimeException("INFERENCE_OOM:${loadedModelId ?: "unknown"}:Out of memory during inference")
@@ -205,14 +283,7 @@ class GemmaInferenceClient(private val context: Context) {
         val tempFile = File(modelsDir, "$modelId.litertlm.tmp")
 
         try {
-          val connection = URL(url).openConnection() as HttpURLConnection
-          connection.connectTimeout = 30_000
-          connection.readTimeout = 30_000
-          connection.connect()
-
-          if (connection.responseCode != HttpURLConnection.HTTP_OK) {
-            throw IOException("HTTP ${connection.responseCode}: ${connection.responseMessage}")
-          }
+          val connection = openConnectionFollowingRedirects(url)
 
           val totalBytes = connection.contentLengthLong
           var bytesRead = 0L
@@ -312,6 +383,49 @@ class GemmaInferenceClient(private val context: Context) {
     }
   }
 
+  /**
+   * Open an HTTP connection, manually following up to 5 redirects across hosts.
+   *
+   * HttpURLConnection follows redirects by default but only within the same host.
+   * HuggingFace LFS redirects from huggingface.co to cdn-lfs-us-1.huggingface.co,
+   * which is a cross-host redirect that HttpURLConnection silently does NOT follow —
+   * it returns the 302 response as-is (or on some Android versions, returns a small
+   * HTML/error body instead of the actual file). This caused downloaded model files
+   * to contain garbage instead of the real model weights.
+   */
+  private fun openConnectionFollowingRedirects(url: String): HttpURLConnection {
+    var currentUrl = url
+    var redirects = 0
+    while (true) {
+      val connection = URL(currentUrl).openConnection() as HttpURLConnection
+      connection.connectTimeout = 30_000
+      connection.readTimeout = 30_000
+      connection.instanceFollowRedirects = false
+      connection.connect()
+
+      val code = connection.responseCode
+      if (code in listOf(
+          HttpURLConnection.HTTP_MOVED_PERM,
+          HttpURLConnection.HTTP_MOVED_TEMP,
+          HttpURLConnection.HTTP_SEE_OTHER,
+          307, 308
+        )) {
+        val location = connection.getHeaderField("Location")
+          ?: throw IOException("Redirect with no Location header")
+        connection.disconnect()
+        redirects++
+        if (redirects > 5) throw IOException("Too many redirects")
+        currentUrl = location
+        continue
+      }
+
+      if (code != HttpURLConnection.HTTP_OK) {
+        throw IOException("HTTP $code: ${connection.responseMessage}")
+      }
+      return connection
+    }
+  }
+
   private fun computeSha256(file: File): String {
     val digest = MessageDigest.getInstance("SHA-256")
     FileInputStream(file).use { fis ->

package/build/ExpoAiKitModule.d.ts

@@ -13,7 +13,7 @@ export interface ExpoAiKitNativeModule {
     getBuiltInModels(): BuiltInModel[];
     getDownloadableModelStatus(modelId: string): DownloadableModelStatus;
     getDeviceRamBytes(): number;
-    setModel(modelId: string): Promise<void>;
+    setModel(modelId: string, minRamBytes: number, backend: string): Promise<void>;
     getActiveModel(): string;
     unloadModel(): Promise<void>;
     downloadModel(modelId: string, url: string, sha256: string): Promise<void>;

package/build/ExpoAiKitModule.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"ExpoAiKitModule.d.ts","sourceRoot":"","sources":["../src/ExpoAiKitModule.ts"],"names":[],"mappings":"AACA,OAAO,KAAK,EAAE,iBAAiB,EAAE,MAAM,mBAAmB,CAAC;AAC3D,OAAO,EACL,UAAU,EACV,WAAW,EACX,cAAc,EACd,YAAY,EACZ,uBAAuB,EACvB,0BAA0B,EAC1B,qBAAqB,EACtB,MAAM,SAAS,CAAC;AAEjB,MAAM,MAAM,qBAAqB,GAAG;IAClC,aAAa,EAAE,CAAC,KAAK,EAAE,cAAc,KAAK,IAAI,CAAC;IAC/C,kBAAkB,EAAE,CAAC,KAAK,EAAE,0BAA0B,KAAK,IAAI,CAAC;IAChE,kBAAkB,EAAE,CAAC,KAAK,EAAE,qBAAqB,KAAK,IAAI,CAAC;CAC5D,CAAC;AAEF,MAAM,WAAW,qBAAqB;IAEpC,WAAW,IAAI,OAAO,CAAC;IACvB,WAAW,CACT,QAAQ,EAAE,UAAU,EAAE,EACtB,YAAY,EAAE,MAAM,GACnB,OAAO,CAAC,WAAW,CAAC,CAAC;IACxB,cAAc,CACZ,QAAQ,EAAE,UAAU,EAAE,EACtB,YAAY,EAAE,MAAM,EACpB,SAAS,EAAE,MAAM,GAChB,OAAO,CAAC,IAAI,CAAC,CAAC;IACjB,aAAa,CAAC,SAAS,EAAE,MAAM,GAAG,OAAO,CAAC,IAAI,CAAC,CAAC;IAGhD,gBAAgB,IAAI,YAAY,EAAE,CAAC;IACnC,0BAA0B,CAAC,OAAO,EAAE,MAAM,GAAG,uBAAuB,CAAC;IACrE,iBAAiB,IAAI,MAAM,CAAC;IAK5B,QAAQ,CAAC,OAAO,EAAE,MAAM,GAAG,OAAO,CAAC,IAAI,CAAC,CAAC;IACzC,cAAc,IAAI,MAAM,CAAC;IAGzB,WAAW,IAAI,OAAO,CAAC,IAAI,CAAC,CAAC;IAG7B,aAAa,CACX,OAAO,EAAE,MAAM,EACf,GAAG,EAAE,MAAM,EACX,MAAM,EAAE,MAAM,GACb,OAAO,CAAC,IAAI,CAAC,CAAC;IACjB,WAAW,CAAC,OAAO,EAAE,MAAM,GAAG,OAAO,CAAC,IAAI,CAAC,CAAC;IAG5C,WAAW,CAAC,CAAC,SAAS,MAAM,qBAAqB,EAC/C,SAAS,EAAE,CAAC,EACZ,QAAQ,EAAE,qBAAqB,CAAC,CAAC,CAAC,GACjC,iBAAiB,CAAC;CACtB;AAED,QAAA,MAAM,eAAe,uBACoC,CAAC;AAE1D,eAAe,eAAe,CAAC"}
+{"version":3,"file":"ExpoAiKitModule.d.ts","sourceRoot":"","sources":["../src/ExpoAiKitModule.ts"],"names":[],"mappings":"AACA,OAAO,KAAK,EAAE,iBAAiB,EAAE,MAAM,mBAAmB,CAAC;AAC3D,OAAO,EACL,UAAU,EACV,WAAW,EACX,cAAc,EACd,YAAY,EACZ,uBAAuB,EACvB,0BAA0B,EAC1B,qBAAqB,EACtB,MAAM,SAAS,CAAC;AAEjB,MAAM,MAAM,qBAAqB,GAAG;IAClC,aAAa,EAAE,CAAC,KAAK,EAAE,cAAc,KAAK,IAAI,CAAC;IAC/C,kBAAkB,EAAE,CAAC,KAAK,EAAE,0BAA0B,KAAK,IAAI,CAAC;IAChE,kBAAkB,EAAE,CAAC,KAAK,EAAE,qBAAqB,KAAK,IAAI,CAAC;CAC5D,CAAC;AAEF,MAAM,WAAW,qBAAqB;IAEpC,WAAW,IAAI,OAAO,CAAC;IACvB,WAAW,CACT,QAAQ,EAAE,UAAU,EAAE,EACtB,YAAY,EAAE,MAAM,GACnB,OAAO,CAAC,WAAW,CAAC,CAAC;IACxB,cAAc,CACZ,QAAQ,EAAE,UAAU,EAAE,EACtB,YAAY,EAAE,MAAM,EACpB,SAAS,EAAE,MAAM,GAChB,OAAO,CAAC,IAAI,CAAC,CAAC;IACjB,aAAa,CAAC,SAAS,EAAE,MAAM,GAAG,OAAO,CAAC,IAAI,CAAC,CAAC;IAGhD,gBAAgB,IAAI,YAAY,EAAE,CAAC;IACnC,0BAA0B,CAAC,OAAO,EAAE,MAAM,GAAG,uBAAuB,CAAC;IACrE,iBAAiB,IAAI,MAAM,CAAC;IAK5B,QAAQ,CAAC,OAAO,EAAE,MAAM,EAAE,WAAW,EAAE,MAAM,EAAE,OAAO,EAAE,MAAM,GAAG,OAAO,CAAC,IAAI,CAAC,CAAC;IAC/E,cAAc,IAAI,MAAM,CAAC;IAGzB,WAAW,IAAI,OAAO,CAAC,IAAI,CAAC,CAAC;IAG7B,aAAa,CACX,OAAO,EAAE,MAAM,EACf,GAAG,EAAE,MAAM,EACX,MAAM,EAAE,MAAM,GACb,OAAO,CAAC,IAAI,CAAC,CAAC;IACjB,WAAW,CAAC,OAAO,EAAE,MAAM,GAAG,OAAO,CAAC,IAAI,CAAC,CAAC;IAG5C,WAAW,CAAC,CAAC,SAAS,MAAM,qBAAqB,EAC/C,SAAS,EAAE,CAAC,EACZ,QAAQ,EAAE,qBAAqB,CAAC,CAAC,CAAC,GACjC,iBAAiB,CAAC;CACtB;AAED,QAAA,MAAM,eAAe,uBACoC,CAAC;AAE1D,eAAe,eAAe,CAAC"}

package/build/ExpoAiKitModule.js.map

@@ -1 +1 @@
-{"version":3,"file":"ExpoAiKitModule.js","sourceRoot":"","sources":["../src/ExpoAiKitModule.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,mBAAmB,EAAE,MAAM,mBAAmB,CAAC;AA6DxD,MAAM,eAAe,GACnB,mBAAmB,CAAwB,WAAW,CAAC,CAAC;AAE1D,eAAe,eAAe,CAAC","sourcesContent":["import { requireNativeModule } from 'expo-modules-core';\nimport type { EventSubscription } from 'expo-modules-core';\nimport {\n  LLMMessage,\n  LLMResponse,\n  LLMStreamEvent,\n  BuiltInModel,\n  DownloadableModelStatus,\n  ModelDownloadProgressEvent,\n  ModelStateChangeEvent,\n} from './types';\n\nexport type ExpoAiKitModuleEvents = {\n  onStreamToken: (event: LLMStreamEvent) => void;\n  onDownloadProgress: (event: ModelDownloadProgressEvent) => void;\n  onModelStateChange: (event: ModelStateChangeEvent) => void;\n};\n\nexport interface ExpoAiKitNativeModule {\n  // Existing inference API\n  isAvailable(): boolean;\n  sendMessage(\n    messages: LLMMessage[],\n    systemPrompt: string\n  ): Promise<LLMResponse>;\n  startStreaming(\n    messages: LLMMessage[],\n    systemPrompt: string,\n    sessionId: string\n  ): Promise<void>;\n  stopStreaming(sessionId: string): Promise<void>;\n\n  // Model discovery\n  getBuiltInModels(): BuiltInModel[];\n  getDownloadableModelStatus(modelId: string): DownloadableModelStatus;\n  getDeviceRamBytes(): number;\n\n  // Model selection & memory management\n  // setModel is async: switching to a downloadable model loads it into memory.\n  // Auto-unloads the previous downloadable model (only one loaded at a time).\n  setModel(modelId: string): Promise<void>;\n  getActiveModel(): string;\n  // Explicitly free memory from the loaded downloadable model.\n  // Reverts to the platform built-in model.\n  unloadModel(): Promise<void>;\n\n  // Model lifecycle (downloadable models only)\n  downloadModel(\n    modelId: string,\n    url: string,\n    sha256: string\n  ): Promise<void>;\n  deleteModel(modelId: string): Promise<void>;\n\n  // Event subscription\n  addListener<K extends keyof ExpoAiKitModuleEvents>(\n    eventName: K,\n    listener: ExpoAiKitModuleEvents[K]\n  ): EventSubscription;\n}\n\nconst ExpoAiKitModule =\n  requireNativeModule<ExpoAiKitNativeModule>('ExpoAiKit');\n\nexport default ExpoAiKitModule;\n"]}
+{"version":3,"file":"ExpoAiKitModule.js","sourceRoot":"","sources":["../src/ExpoAiKitModule.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,mBAAmB,EAAE,MAAM,mBAAmB,CAAC;AA6DxD,MAAM,eAAe,GACnB,mBAAmB,CAAwB,WAAW,CAAC,CAAC;AAE1D,eAAe,eAAe,CAAC","sourcesContent":["import { requireNativeModule } from 'expo-modules-core';\nimport type { EventSubscription } from 'expo-modules-core';\nimport {\n  LLMMessage,\n  LLMResponse,\n  LLMStreamEvent,\n  BuiltInModel,\n  DownloadableModelStatus,\n  ModelDownloadProgressEvent,\n  ModelStateChangeEvent,\n} from './types';\n\nexport type ExpoAiKitModuleEvents = {\n  onStreamToken: (event: LLMStreamEvent) => void;\n  onDownloadProgress: (event: ModelDownloadProgressEvent) => void;\n  onModelStateChange: (event: ModelStateChangeEvent) => void;\n};\n\nexport interface ExpoAiKitNativeModule {\n  // Existing inference API\n  isAvailable(): boolean;\n  sendMessage(\n    messages: LLMMessage[],\n    systemPrompt: string\n  ): Promise<LLMResponse>;\n  startStreaming(\n    messages: LLMMessage[],\n    systemPrompt: string,\n    sessionId: string\n  ): Promise<void>;\n  stopStreaming(sessionId: string): Promise<void>;\n\n  // Model discovery\n  getBuiltInModels(): BuiltInModel[];\n  getDownloadableModelStatus(modelId: string): DownloadableModelStatus;\n  getDeviceRamBytes(): number;\n\n  // Model selection & memory management\n  // setModel is async: switching to a downloadable model loads it into memory.\n  // Auto-unloads the previous downloadable model (only one loaded at a time).\n  setModel(modelId: string, minRamBytes: number, backend: string): Promise<void>;\n  getActiveModel(): string;\n  // Explicitly free memory from the loaded downloadable model.\n  // Reverts to the platform built-in model.\n  unloadModel(): Promise<void>;\n\n  // Model lifecycle (downloadable models only)\n  downloadModel(\n    modelId: string,\n    url: string,\n    sha256: string\n  ): Promise<void>;\n  deleteModel(modelId: string): Promise<void>;\n\n  // Event subscription\n  addListener<K extends keyof ExpoAiKitModuleEvents>(\n    eventName: K,\n    listener: ExpoAiKitModuleEvents[K]\n  ): EventSubscription;\n}\n\nconst ExpoAiKitModule =\n  requireNativeModule<ExpoAiKitNativeModule>('ExpoAiKit');\n\nexport default ExpoAiKitModule;\n"]}

package/build/index.d.ts

@@ -1,4 +1,4 @@
-import { LLMMessage, LLMSendOptions, LLMResponse, LLMStreamOptions, LLMStreamCallback, LLMSummarizeOptions, LLMTranslateOptions, LLMRewriteOptions, LLMExtractKeyPointsOptions, LLMAnswerQuestionOptions, LLMSuggestOptions, LLMSuggestResponse, LLMSmartReplyOptions, LLMAutocompleteOptions, BuiltInModel, DownloadableModel } from './types';
+import { LLMMessage, LLMSendOptions, LLMResponse, LLMStreamOptions, LLMStreamCallback, LLMSummarizeOptions, LLMTranslateOptions, LLMRewriteOptions, LLMExtractKeyPointsOptions, LLMAnswerQuestionOptions, LLMSuggestOptions, LLMSuggestResponse, LLMSmartReplyOptions, LLMAutocompleteOptions, BuiltInModel, DownloadableModel, SetModelOptions } from './types';
 export * from './types';
 export * from './memory';
 export * from './hooks';
@@ -575,12 +575,14 @@ export declare function deleteModel(modelId: string): Promise<void>;
  * model (today's behavior, no error).
  *
  * @param modelId - ID of the model to activate (e.g. 'gemma-e2b', 'apple-fm', 'mlkit')
+ * @param options - Optional configuration for model loading
+ * @param options.backend - Hardware backend: 'auto' (default, GPU with CPU fallback), 'gpu', or 'cpu'
  * @throws {ModelError} MODEL_NOT_FOUND if modelId is invalid
  * @throws {ModelError} MODEL_NOT_DOWNLOADED if the downloadable model file is not on disk
  * @throws {ModelError} MODEL_LOAD_FAILED if loading into memory fails
  * @throws {ModelError} INFERENCE_OOM if device can't fit model in memory
  */
-export declare function setModel(modelId: string): Promise<void>;
+export declare function setModel(modelId: string, options?: SetModelOptions): Promise<void>;
 /**
  * Get the ID of the currently active model.
  *

package/build/index.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":"AAEA,OAAO,EACL,UAAU,EACV,cAAc,EACd,WAAW,EACX,gBAAgB,EAEhB,iBAAiB,EACjB,mBAAmB,EACnB,mBAAmB,EACnB,iBAAiB,EACjB,0BAA0B,EAC1B,wBAAwB,EACxB,iBAAiB,EACjB,kBAAkB,EAClB,oBAAoB,EACpB,sBAAsB,EACtB,YAAY,EACZ,iBAAiB,EAElB,MAAM,SAAS,CAAC;AAGjB,cAAc,SAAS,CAAC;AACxB,cAAc,UAAU,CAAC;AACzB,cAAc,SAAS,CAAC;AACxB,cAAc,UAAU,CAAC;AAiJzB;;;GAGG;AACH,wBAAsB,WAAW,IAAI,OAAO,CAAC,OAAO,CAAC,CAKpD;AAED;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GAkCG;AACH,wBAAsB,WAAW,CAC/B,QAAQ,EAAE,UAAU,EAAE,EACtB,OAAO,CAAC,EAAE,cAAc,GACvB,OAAO,CAAC,WAAW,CAAC,CAgBtB;AAED;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GA2CG;AACH,wBAAgB,aAAa,CAC3B,QAAQ,EAAE,UAAU,EAAE,EACtB,OAAO,EAAE,iBAAiB,EAC1B,OAAO,CAAC,EAAE,gBAAgB,GACzB;IAAE,OAAO,EAAE,OAAO,CAAC,WAAW,CAAC,CAAC;IAAC,IAAI,EAAE,MAAM,IAAI,CAAA;CAAE,CAiErD;AAMD;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GA8BG;AACH,wBAAsB,SAAS,CAC7B,IAAI,EAAE,MAAM,EACZ,OAAO,CAAC,EAAE,mBAAmB,GAC5B,OAAO,CAAC,WAAW,CAAC,CActB;AAED;;;;;;;;;;;;;;;;;GAiBG;AACH,wBAAgB,eAAe,CAC7B,IAAI,EAAE,MAAM,EACZ,OAAO,EAAE,iBAAiB,EAC1B,OAAO,CAAC,EAAE,mBAAmB,GAC5B;IAAE,OAAO,EAAE,OAAO,CAAC,WAAW,CAAC,CAAC;IAAC,IAAI,EAAE,MAAM,IAAI,CAAA;CAAE,CAmBrD;AAED;;;;;;;;;;;;;;;;;;;;;;;GAuBG;AACH,wBAAsB,SAAS,CAC7B,IAAI,EAAE,MAAM,EACZ,OAAO,EAAE,mBAAmB,GAC3B,OAAO,CAAC,WAAW,CAAC,CAatB;AAED;;;;;;;;;;;;;;;;;GAiBG;AACH,wBAAgB,eAAe,CAC7B,IAAI,EAAE,MAAM,EACZ,OAAO,EAAE,iBAAiB,EAC1B,OAAO,EAAE,mBAAmB,GAC3B;IAAE,OAAO,EAAE,OAAO,CAAC,WAAW,CAAC,CAAC;IAAC,IAAI,EAAE,MAAM,IAAI,CAAA;CAAE,CAkBrD;AAED;;;;;;;;;;;;;;;;;;;;;GAqBG;AACH,wBAAsB,OAAO,CAC3B,IAAI,EAAE,MAAM,EACZ,OAAO,EAAE,iBAAiB,GACzB,OAAO,CAAC,WAAW,CAAC,CAatB;AAED;;;;;;;;;;;;;;;;;GAiBG;AACH,wBAAgB,aAAa,CAC3B,IAAI,EAAE,MAAM,EACZ,OAAO,EAAE,iBAAiB,EAC1B,OAAO,EAAE,iBAAiB,GACzB;IAAE,OAAO,EAAE,OAAO,CAAC,WAAW,CAAC,CAAC;IAAC,IAAI,EAAE,MAAM,IAAI,CAAA;CAAE,CAkBrD;AAED;;;;;;;;;;;;;;;;;;;;GAoBG;AACH,wBAAsB,gBAAgB,CACpC,IAAI,EAAE,MAAM,EACZ,OAAO,CAAC,EAAE,0BAA0B,GACnC,OAAO,CAAC,WAAW,CAAC,CAatB;AAED;;;;;;;;;;;;;;;;;GAiBG;AACH,wBAAgB,sBAAsB,CACpC,IAAI,EAAE,MAAM,EACZ,OAAO,EAAE,iBAAiB,EAC1B,OAAO,CAAC,EAAE,0BAA0B,GACnC;IAAE,OAAO,EAAE,OAAO,CAAC,WAAW,CAAC,CAAC;IAAC,IAAI,EAAE,MAAM,IAAI,CAAA;CAAE,CAkBrD;AAED;;;;;;;;;;;;;;;;;;;;;;;;;;;GA2BG;AACH,wBAAsB,cAAc,CAClC,QAAQ,EAAE,MAAM,EAChB,OAAO,EAAE,MAAM,EACf,OAAO,CAAC,EAAE,wBAAwB,GACjC,OAAO,CAAC,WAAW,CAAC,CAkBtB;AAED;;;;;;;;;;;;;;;;;;;GAmBG;AACH,wBAAgB,oBAAoB,CAClC,QAAQ,EAAE,MAAM,EAChB,OAAO,EAAE,MAAM,EACf,OAAO,EAAE,iBAAiB,EAC1B,OAAO,CAAC,EAAE,wBAAwB,GACjC;IAAE,OAAO,EAAE,OAAO,CAAC,WAAW,CAAC,CAAC;IAAC,IAAI,EAAE,MAAM,IAAI,CAAA;CAAE,CA0BrD;AAMD;;;;;;;;;;;;;;;;;;;;;;;;;;;;GA4BG;AACH,wBAAsB,OAAO,CAC3B,WAAW,EAAE,MAAM,EACnB,OAAO,CAAC,EAAE,iBAAiB,GAC1B,OAAO,CAAC,kBAAkB,CAAC,CAsB7B;AAED;;;;;;;;;;;;;;;;;GAiBG;AACH,wBAAgB,aAAa,CAC3B,WAAW,EAAE,MAAM,EACnB,OAAO,EAAE,iBAAiB,EAC1B,OAAO,CAAC,EAAE,iBAAiB,GAC1B;IAAE,OAAO,EAAE,OAAO,CAAC,WAAW,CAAC,CAAC;IAAC,IAAI,EAAE,MAAM,IAAI,CAAA;CAAE,CAmBrD;AAED;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GAkCG;AACH,wBAAsB,UAAU,CAC9B,QAAQ,EAAE,UAAU,EAAE,EACtB,OAAO,CAAC,EAAE,oBAAoB,GAC7B,OAAO,CAAC,kBAAkB,CAAC,CA0B7B;AAED;;;;;;;;;;;;;;;;GAgBG;AACH,wBAAgB,gBAAgB,CAC9B,QAAQ,EAAE,UAAU,EAAE,EACtB,OAAO,EAAE,iBAAiB,EAC1B,OAAO,CAAC,EAAE,oBAAoB,GAC7B;IAAE,OAAO,EAAE,OAAO,CAAC,WAAW,CAAC,CAAC;IAAC,IAAI,EAAE,MAAM,IAAI,CAAA;CAAE,CAyBrD;AAED;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GA6BG;AACH,wBAAsB,YAAY,CAChC,WAAW,EAAE,MAAM,EACnB,OAAO,CAAC,EAAE,sBAAsB,GAC/B,OAAO,CAAC,kBAAkB,CAAC,CAsB7B;AAED;;;;;;;;;;;;;;;;GAgBG;AACH,wBAAgB,kBAAkB,CAChC,WAAW,EAAE,MAAM,EACnB,OAAO,EAAE,iBAAiB,EAC1B,OAAO,CAAC,EAAE,sBAAsB,GAC/B;IAAE,OAAO,EAAE,OAAO,CAAC,WAAW,CAAC,CAAC;IAAC,IAAI,EAAE,MAAM,IAAI,CAAA;CAAE,CAmBrD;AAED;;;;;;;;;;;;;;;;GAgBG;AACH,wBAAgB,oBAAoB,CAAC,GAAG,EAAE,MAAM,GAAG,kBAAkB,CAKpE;AAMD;;;;;;;GAOG;AACH,wBAAsB,gBAAgB,IAAI,OAAO,CAAC,YAAY,EAAE,CAAC,CAKhE;AAED;;;;;;;GAOG;AACH,wBAAsB,qBAAqB,IAAI,OAAO,CAAC,iBAAiB,EAAE,CAAC,CA6B1E;AAED;;;;;;;;;;;;;;GAcG;AACH,wBAAsB,aAAa,CACjC,OAAO,EAAE,MAAM,EACf,OAAO,CAAC,EAAE;IAAE,UAAU,CAAC,EAAE,CAAC,QAAQ,EAAE,MAAM,KAAK,IAAI,CAAA;CAAE,GACpD,OAAO,CAAC,IAAI,CAAC,CAiDf;AAED;;;;;;;GAOG;AACH,wBAAsB,WAAW,CAAC,OAAO,EAAE,MAAM,GAAG,OAAO,CAAC,IAAI,CAAC,CAOhE;AAED;;;;;;;;;;;;;;;;;;;;GAoBG;AACH,wBAAsB,QAAQ,CAAC,OAAO,EAAE,MAAM,GAAG,OAAO,CAAC,IAAI,CAAC,CAE7D;AAED;;;;GAIG;AACH,wBAAgB,cAAc,IAAI,MAAM,CAEvC;AAED;;;;;GAKG;AACH,wBAAsB,WAAW,IAAI,OAAO,CAAC,IAAI,CAAC,CAEjD"}
+{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":"AAEA,OAAO,EACL,UAAU,EACV,cAAc,EACd,WAAW,EACX,gBAAgB,EAEhB,iBAAiB,EACjB,mBAAmB,EACnB,mBAAmB,EACnB,iBAAiB,EACjB,0BAA0B,EAC1B,wBAAwB,EACxB,iBAAiB,EACjB,kBAAkB,EAClB,oBAAoB,EACpB,sBAAsB,EACtB,YAAY,EACZ,iBAAiB,EAEjB,eAAe,EAChB,MAAM,SAAS,CAAC;AAGjB,cAAc,SAAS,CAAC;AACxB,cAAc,UAAU,CAAC;AACzB,cAAc,SAAS,CAAC;AACxB,cAAc,UAAU,CAAC;AAiJzB;;;GAGG;AACH,wBAAsB,WAAW,IAAI,OAAO,CAAC,OAAO,CAAC,CAKpD;AAED;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GAkCG;AACH,wBAAsB,WAAW,CAC/B,QAAQ,EAAE,UAAU,EAAE,EACtB,OAAO,CAAC,EAAE,cAAc,GACvB,OAAO,CAAC,WAAW,CAAC,CAgBtB;AAED;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GA2CG;AACH,wBAAgB,aAAa,CAC3B,QAAQ,EAAE,UAAU,EAAE,EACtB,OAAO,EAAE,iBAAiB,EAC1B,OAAO,CAAC,EAAE,gBAAgB,GACzB;IAAE,OAAO,EAAE,OAAO,CAAC,WAAW,CAAC,CAAC;IAAC,IAAI,EAAE,MAAM,IAAI,CAAA;CAAE,CAiErD;AAMD;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GA8BG;AACH,wBAAsB,SAAS,CAC7B,IAAI,EAAE,MAAM,EACZ,OAAO,CAAC,EAAE,mBAAmB,GAC5B,OAAO,CAAC,WAAW,CAAC,CActB;AAED;;;;;;;;;;;;;;;;;GAiBG;AACH,wBAAgB,eAAe,CAC7B,IAAI,EAAE,MAAM,EACZ,OAAO,EAAE,iBAAiB,EAC1B,OAAO,CAAC,EAAE,mBAAmB,GAC5B;IAAE,OAAO,EAAE,OAAO,CAAC,WAAW,CAAC,CAAC;IAAC,IAAI,EAAE,MAAM,IAAI,CAAA;CAAE,CAmBrD;AAED;;;;;;;;;;;;;;;;;;;;;;;GAuBG;AACH,wBAAsB,SAAS,CAC7B,IAAI,EAAE,MAAM,EACZ,OAAO,EAAE,mBAAmB,GAC3B,OAAO,CAAC,WAAW,CAAC,CAatB;AAED;;;;;;;;;;;;;;;;;GAiBG;AACH,wBAAgB,eAAe,CAC7B,IAAI,EAAE,MAAM,EACZ,OAAO,EAAE,iBAAiB,EAC1B,OAAO,EAAE,mBAAmB,GAC3B;IAAE,OAAO,EAAE,OAAO,CAAC,WAAW,CAAC,CAAC;IAAC,IAAI,EAAE,MAAM,IAAI,CAAA;CAAE,CAkBrD;AAED;;;;;;;;;;;;;;;;;;;;;GAqBG;AACH,wBAAsB,OAAO,CAC3B,IAAI,EAAE,MAAM,EACZ,OAAO,EAAE,iBAAiB,GACzB,OAAO,CAAC,WAAW,CAAC,CAatB;AAED;;;;;;;;;;;;;;;;;GAiBG;AACH,wBAAgB,aAAa,CAC3B,IAAI,EAAE,MAAM,EACZ,OAAO,EAAE,iBAAiB,EAC1B,OAAO,EAAE,iBAAiB,GACzB;IAAE,OAAO,EAAE,OAAO,CAAC,WAAW,CAAC,CAAC;IAAC,IAAI,EAAE,MAAM,IAAI,CAAA;CAAE,CAkBrD;AAED;;;;;;;;;;;;;;;;;;;;GAoBG;AACH,wBAAsB,gBAAgB,CACpC,IAAI,EAAE,MAAM,EACZ,OAAO,CAAC,EAAE,0BAA0B,GACnC,OAAO,CAAC,WAAW,CAAC,CAatB;AAED;;;;;;;;;;;;;;;;;GAiBG;AACH,wBAAgB,sBAAsB,CACpC,IAAI,EAAE,MAAM,EACZ,OAAO,EAAE,iBAAiB,EAC1B,OAAO,CAAC,EAAE,0BAA0B,GACnC;IAAE,OAAO,EAAE,OAAO,CAAC,WAAW,CAAC,CAAC;IAAC,IAAI,EAAE,MAAM,IAAI,CAAA;CAAE,CAkBrD;AAED;;;;;;;;;;;;;;;;;;;;;;;;;;;GA2BG;AACH,wBAAsB,cAAc,CAClC,QAAQ,EAAE,MAAM,EAChB,OAAO,EAAE,MAAM,EACf,OAAO,CAAC,EAAE,wBAAwB,GACjC,OAAO,CAAC,WAAW,CAAC,CAkBtB;AAED;;;;;;;;;;;;;;;;;;;GAmBG;AACH,wBAAgB,oBAAoB,CAClC,QAAQ,EAAE,MAAM,EAChB,OAAO,EAAE,MAAM,EACf,OAAO,EAAE,iBAAiB,EAC1B,OAAO,CAAC,EAAE,wBAAwB,GACjC;IAAE,OAAO,EAAE,OAAO,CAAC,WAAW,CAAC,CAAC;IAAC,IAAI,EAAE,MAAM,IAAI,CAAA;CAAE,CA0BrD;AAMD;;;;;;;;;;;;;;;;;;;;;;;;;;;;GA4BG;AACH,wBAAsB,OAAO,CAC3B,WAAW,EAAE,MAAM,EACnB,OAAO,CAAC,EAAE,iBAAiB,GAC1B,OAAO,CAAC,kBAAkB,CAAC,CAsB7B;AAED;;;;;;;;;;;;;;;;;GAiBG;AACH,wBAAgB,aAAa,CAC3B,WAAW,EAAE,MAAM,EACnB,OAAO,EAAE,iBAAiB,EAC1B,OAAO,CAAC,EAAE,iBAAiB,GAC1B;IAAE,OAAO,EAAE,OAAO,CAAC,WAAW,CAAC,CAAC;IAAC,IAAI,EAAE,MAAM,IAAI,CAAA;CAAE,CAmBrD;AAED;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GAkCG;AACH,wBAAsB,UAAU,CAC9B,QAAQ,EAAE,UAAU,EAAE,EACtB,OAAO,CAAC,EAAE,oBAAoB,GAC7B,OAAO,CAAC,kBAAkB,CAAC,CA0B7B;AAED;;;;;;;;;;;;;;;;GAgBG;AACH,wBAAgB,gBAAgB,CAC9B,QAAQ,EAAE,UAAU,EAAE,EACtB,OAAO,EAAE,iBAAiB,EAC1B,OAAO,CAAC,EAAE,oBAAoB,GAC7B;IAAE,OAAO,EAAE,OAAO,CAAC,WAAW,CAAC,CAAC;IAAC,IAAI,EAAE,MAAM,IAAI,CAAA;CAAE,CAyBrD;AAED;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GA6BG;AACH,wBAAsB,YAAY,CAChC,WAAW,EAAE,MAAM,EACnB,OAAO,CAAC,EAAE,sBAAsB,GAC/B,OAAO,CAAC,kBAAkB,CAAC,CAsB7B;AAED;;;;;;;;;;;;;;;;GAgBG;AACH,wBAAgB,kBAAkB,CAChC,WAAW,EAAE,MAAM,EACnB,OAAO,EAAE,iBAAiB,EAC1B,OAAO,CAAC,EAAE,sBAAsB,GAC/B;IAAE,OAAO,EAAE,OAAO,CAAC,WAAW,CAAC,CAAC;IAAC,IAAI,EAAE,MAAM,IAAI,CAAA;CAAE,CAmBrD;AAED;;;;;;;;;;;;;;;;GAgBG;AACH,wBAAgB,oBAAoB,CAAC,GAAG,EAAE,MAAM,GAAG,kBAAkB,CAKpE;AAMD;;;;;;;GAOG;AACH,wBAAsB,gBAAgB,IAAI,OAAO,CAAC,YAAY,EAAE,CAAC,CAKhE;AAED;;;;;;;GAOG;AACH,wBAAsB,qBAAqB,IAAI,OAAO,CAAC,iBAAiB,EAAE,CAAC,CA6B1E;AAED;;;;;;;;;;;;;;GAcG;AACH,wBAAsB,aAAa,CACjC,OAAO,EAAE,MAAM,EACf,OAAO,CAAC,EAAE;IAAE,UAAU,CAAC,EAAE,CAAC,QAAQ,EAAE,MAAM,KAAK,IAAI,CAAA;CAAE,GACpD,OAAO,CAAC,IAAI,CAAC,CAiDf;AAED;;;;;;;GAOG;AACH,wBAAsB,WAAW,CAAC,OAAO,EAAE,MAAM,GAAG,OAAO,CAAC,IAAI,CAAC,CAOhE;AAED;;;;;;;;;;;;;;;;;;;;;;GAsBG;AACH,wBAAsB,QAAQ,CAAC,OAAO,EAAE,MAAM,EAAE,OAAO,CAAC,EAAE,eAAe,GAAG,OAAO,CAAC,IAAI,CAAC,CAKxF;AAED;;;;GAIG;AACH,wBAAgB,cAAc,IAAI,MAAM,CAEvC;AAED;;;;;GAKG;AACH,wBAAsB,WAAW,IAAI,OAAO,CAAC,IAAI,CAAC,CAEjD"}

package/build/index.js

@@ -1024,13 +1024,18 @@ export async function deleteModel(modelId) {
  * model (today's behavior, no error).
  *
  * @param modelId - ID of the model to activate (e.g. 'gemma-e2b', 'apple-fm', 'mlkit')
+ * @param options - Optional configuration for model loading
+ * @param options.backend - Hardware backend: 'auto' (default, GPU with CPU fallback), 'gpu', or 'cpu'
  * @throws {ModelError} MODEL_NOT_FOUND if modelId is invalid
  * @throws {ModelError} MODEL_NOT_DOWNLOADED if the downloadable model file is not on disk
  * @throws {ModelError} MODEL_LOAD_FAILED if loading into memory fails
  * @throws {ModelError} INFERENCE_OOM if device can't fit model in memory
  */
-export async function setModel(modelId) {
-    await ExpoAiKitModule.setModel(modelId);
+export async function setModel(modelId, options) {
+    const entry = getRegistryEntry(modelId);
+    const minRamBytes = entry?.minRamBytes ?? 0;
+    const backend = options?.backend ?? 'auto';
+    await ExpoAiKitModule.setModel(modelId, minRamBytes, backend);
 }
 /**
  * Get the ID of the currently active model.