Update app.py

app.py

@@ -24,7 +24,7 @@ INITIAL_MESSAGE = """Halo! 👋 Saya adalah asisten kesehatan feminacare yang si
 Silakan ajukan pertanyaan apa saja dan saya akan membantu Anda dengan informasi yang akurat."""
 
 # Model configurations
-MODEL_NAME = "SeaLLMs/SeaLLMs-v3-7B-Chat"
+MODEL_NAME = "PrunaAI/SeaLLMs-SeaLLM-7B-v2.5-QUANTO-int2bit-smashed"
 EMBEDDING_MODEL = "sentence-transformers/all-MiniLM-L6-v2"
 TOP_K_DOCS = 5
 
@@ -67,25 +67,25 @@ def create_llm():
     
     # llm = HuggingFacePipeline(pipeline=text_generation_pipeline)
 
-    # return HuggingFaceHub(
-    #     repo_id=MODEL_NAME,
-    #     model_kwargs={
-    #         "temperature": 0.7,  # Balanced between creativity and accuracy
-    #         "max_new_tokens": 1024,
-    #         "top_p": 0.9,
-    #         "frequency_penalty": 0.5
-    #     }
-    # )
-    llm = HuggingFacePipeline.from_model_id(
-        model_id=MODEL_NAME,
-        task="text-generation",
-        pipeline_kwargs=dict(
-            max_new_tokens=512,
-            do_sample=False,
-            repetition_penalty=1.03,
-        ),
+    return HuggingFaceHub(
+        repo_id=MODEL_NAME,
+        model_kwargs={
+            "temperature": 0.7,  # Balanced between creativity and accuracy
+            "max_new_tokens": 1024,
+            "top_p": 0.9,
+            "frequency_penalty": 0.5
+        }
     )
-    return llm
+    # llm = HuggingFacePipeline.from_model_id(
+    #     model_id=MODEL_NAME,
+    #     task="text-generation",
+    #     pipeline_kwargs=dict(
+    #         max_new_tokens=512,
+    #         do_sample=False,
+    #         repetition_penalty=1.03,
+    #     ),
+    # )
+    # return llm
     
     
     # chat_model = ChatHuggingFace(llm=llm)