Update app.py

app.py

@@ -9,7 +9,7 @@ personality_model = AutoModelForSequenceClassification.from_pretrained("KevSun/P
 personality_tokenizer = AutoTokenizer.from_pretrained("KevSun/Personality_LM")
 
 # Initialize the LLM client (HuggingFaceH4/zephyr-7b-beta)
-llm_client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
+llm_client = InferenceClient("Qwen/Qwen2.5-72B-Instruct")
 
 def analyze_personality(text):
     """
@@ -70,9 +70,9 @@ def generate_response(user_message, traits):
     response = ""
     for message in llm_client.chat_completion(
         messages,
-        max_tokens=256,
+        max_tokens=100,
         stream=True,
-        temperature=0.7,
+        temperature=0.9,
         top_p=0.95
     ):
         token = message.choices[0].delta.content