[Android] Add API to use new config

Differential Revision: D73556369

Pull Request resolved: #10346

Author: kirklandsign

extension/android/executorch_android/src/main/java/org/pytorch/executorch/extension/llm/LlmModule.java

@@ -59,6 +59,17 @@ public LlmModule(int modelType, String modulePath, String tokenizerPath, float t
     mHybridData = initHybrid(modelType, modulePath, tokenizerPath, temperature, null);
   }
 
+  /** Constructs a LLM Module for a model with the given LlmModuleConfig */
+  public LlmModule(LlmModuleConfig config) {
+    mHybridData =
+        initHybrid(
+            config.getModelType(),
+            config.getModulePath(),
+            config.getTokenizerPath(),
+            config.getTemperature(),
+            config.getDataPath());
+  }
+
   public void resetNative() {
     mHybridData.resetNative();
   }
@@ -107,6 +118,19 @@ public int generate(String prompt, int seqLen, LlmCallback llmCallback, boolean
     return generate(null, 0, 0, 0, prompt, seqLen, llmCallback, echo);
   }
 
+  /**
+   * Start generating tokens from the module.
+   *
+   * @param prompt Input prompt
+   * @param config the config for generation
+   * @param llmCallback callback object to receive results
+   */
+  public int generate(String prompt, LlmGenerationConfig config, LlmCallback llmCallback) {
+    int seqLen = config.getSeqLen();
+    boolean echo = config.isEcho();
+    return generate(null, 0, 0, 0, prompt, seqLen, llmCallback, echo);
+  }
+
   /**
    * Start generating tokens from the module.
    *