android sdk upgrade

android/llama.android/.gitignore → android/.gitignore

@@ -2,6 +2,8 @@
 .gradle/
 build/
 
+.idea
+
 # Local configuration file (sdk path, etc)
 local.properties
 

android/llama.android/README.md → android/README.md

@@ -51,4 +51,9 @@ Open the [android test project](./app-java) folder in Android Studio and run the
 
 ## Download Models
 
-You can download models from the [Nexa AI ModelHub](https://nexa.ai/models).
+You can download models from the [Nexa AI ModelHub](https://nexa.ai/models).
+
+## How to estimate power usage
+
+- ```adb shell dumpsys batterystats --reset```
+- ```adb shell dumpsys batterystats > batterystats.txt```

android/llama.android/app-java/src/main/AndroidManifest.xml → android/app-java/src/main/AndroidManifest.xml

@@ -1,6 +1,6 @@
 <?xml version="1.0" encoding="utf-8"?>
-<manifest xmlns:android="http://schemas.android.com/apk/res/android">
-
+<manifest xmlns:android="http://schemas.android.com/apk/res/android"
+    xmlns:tools="http://schemas.android.com/tools">
     <uses-permission android:name="android.permission.ACCESS_NETWORK_STATE" />
     <uses-permission android:name="android.permission.INTERNET" />
     <uses-permission android:name="android.permission.RECORD_AUDIO"/>
@@ -12,6 +12,9 @@
     <uses-permission android:name="android.permission.ACCESS_COARSE_LOCATION" />
 
     <application
+        android:largeHeap="true"
+        android:hardwareAccelerated="true"
+        tools:replace="android:largeHeap"
         android:allowBackup="true"
         android:dataExtractionRules="@xml/data_extraction_rules"
         android:fullBackupContent="@xml/backup_rules"

android/llama.android/app-java/src/main/java/ai/nexa/app_java/LlamaBridge.java → android/app-java/src/main/java/ai/nexa/app_java/LlamaBridge.java

@@ -1,7 +1,7 @@
 package ai.nexa.app_java;
 
 import android.content.Context;
-import com.nexa.NexaVlmInference;
+import com.nexa.NexaOmniVlmInference;
 import android.util.Log;
 
 import java.io.IOException;
@@ -29,7 +29,7 @@ public class LlamaBridge {
     private final MessageHandler messageHandler;
     private final VlmModelManager modelManager;
     private final ImagePathHelper imagePathHelper;
-    private NexaVlmInference nexaVlmInference;
+    private NexaOmniVlmInference nexaVlmInference;
     private boolean isModelLoaded = false;
 
     private final KotlinFlowHelper flowHelper = new KotlinFlowHelper();
@@ -73,7 +73,7 @@ public void loadModel() {
                 Log.d(TAG, "Loading projector from: " + projectorPath);
 
                 // Create with default values for optional parameters
-                nexaVlmInference = new NexaVlmInference(
+                nexaVlmInference = new NexaOmniVlmInference(
                         modelPath,            // modelPath
                         projectorPath,        // projectorPath
                         "",                   // imagePath (empty string as default)

android/llama.android/app-java/src/main/java/ai/nexa/app_java/VlmModelManager.java → android/app-java/src/main/java/ai/nexa/app_java/VlmModelManager.java

@@ -10,8 +10,15 @@
 public class VlmModelManager {
     private static final String TAG = "LlamaBridge";
     private static final String MODELS_DIR = "models";
-    private static final String MODEL_TEXT_FILENAME = "nanollava-text-model-q4_0.gguf";
-    private static final String MODEL_MMPROJ_FILENAME = "nanollava-mmproj-f16.gguf";
+//    private static final String MODEL_TEXT_FILENAME = "nanollava-text-model-q4_0.gguf";
+//    private static final String MODEL_MMPROJ_FILENAME = "nanollava-mmproj-f16.gguf";
+
+    private static final String MODEL_TEXT_FILENAME = "model-q8_0.gguf";
+    private static final String MODEL_MMPROJ_FILENAME = "projector-q8_0.gguf";
+
+//    private static final String MODEL_TEXT_FILENAME = "nano-vlm-instruct-llm-F16.gguf";
+//    private static final String MODEL_MMPROJ_FILENAME = "nano-vlm-instruct-mmproj-F16.gguf";
+
 
     private final Context context;
     private File textModelFile;

android/llama.android/gradle.properties → android/gradle.properties

@@ -6,7 +6,7 @@
 # http://www.gradle.org/docs/current/userguide/build_environment.html
 # Specifies the JVM arguments used for the daemon process.
 # The setting is particularly useful for tweaking memory settings.
-org.gradle.jvmargs=-Xmx2048m -Dfile.encoding=UTF-8
+org.gradle.jvmargs=-Xms1024m -Xmx4g -XX:MaxDirectMemorySize=3g -Dsun.nio.MaxDirectMemorySize=3g -Dfile.encoding=UTF-8 -Dorg.gradle.parallel=true -Dorg.gradle.workers.max=4
 # When configured, Gradle will run in incubating parallel mode.
 # This option should only be used with decoupled projects. More details, visit
 # http://www.gradle.org/docs/current/userguide/multi_project_builds.html#sec:decoupled_projects
@@ -21,3 +21,6 @@ kotlin.code.style=official
 # resources declared in the library itself and none from the library's dependencies,
 # thereby reducing the size of the R class for that library
 android.nonTransitiveRClass=true
+
+android.enableR8.fullMode=true
+android.debug.obsoleteApi=true