Update app.py

app.py

@@ -22,20 +22,15 @@ from transformers import AutoModelForCausalLM, AutoTokenizer
 
 device = "cpu" # for GPU usage or "cpu" for CPU usage
 
-tokenizer0 = AutoTokenizer.from_pretrained(MODEL_LIST[0])
-model0 = AutoModelForCausalLM.from_pretrained(MODEL_LIST[0]).to(device)
+tokenizer = AutoTokenizer.from_pretrained(MODEL_LIST[0])
+model = AutoModelForCausalLM.from_pretrained(MODEL_LIST[0]).to(device)
 
-tokenizer1 = AutoTokenizer.from_pretrained(MODEL_LIST[1])
-model1 = AutoModelForCausalLM.from_pretrained(MODEL_LIST[1]).to(device)
-
-tokenizer2 = AutoTokenizer.from_pretrained(MODEL_LIST[2])
-model2 = AutoModelForCausalLM.from_pretrained(MODEL_LIST[2]).to(device)
 
 #@spaces.GPU()
 def stream_chat(
     message: str, 
     history: list, 
-    temperature: float = 0.8, 
+    temperature: float = 0.4, 
     max_new_tokens: int = 1024, 
     top_p: float = 1.0, 
     top_k: int = 20, 
@@ -55,8 +50,6 @@ def stream_chat(
     conversation.append({"role": "user", "content": message})
 
     
-    model = model2
-    tokenizer = tokenizer2
 
     input_text=tokenizer.apply_chat_template(conversation,  add_generation_prompt=True, tokenize=False)
     inputs = tokenizer.encode(input_text, return_tensors="pt").to(device)