Update model loading with English language support and system prompt

- Replace Spanish print statements with English equivalents
- Add a system prompt to the model message generation
- Improve language handling for the chatbot

app.py

@@ -21,7 +21,7 @@ try:
     tokenizer = AutoTokenizer.from_pretrained(model_name)
     
     if device == "cuda":
-        print("Usando GPU para el modelo...")
+        print("Using GPU for the model...")
         model = AutoModelForCausalLM.from_pretrained(
             model_name,
             torch_dtype=torch.bfloat16,
@@ -29,16 +29,16 @@ try:
             low_cpu_mem_usage=True
         )
     else:
-        print("Usando CPU para el modelo...")
+        print("Using CPU for the model...")
         model = AutoModelForCausalLM.from_pretrained(
             model_name,
             device_map={"": device},
             torch_dtype=torch.float32
         )
 
-    print(f"Modelo cargado exitosamente en: {device}")
+    print(f"Model loaded successfully on: {device}")
 except Exception as e:
-    print(f"Error al cargar el modelo: {str(e)}")
+    print(f"Error loading the model: {str(e)}")
     raise
 
 # Define the function that calls the model
@@ -53,7 +53,10 @@ def call_model(state: MessagesState):
         dict: A dictionary containing the generated text and the thread ID
     """
     # Convert LangChain messages to chat format
-    messages = []
+    messages = [
+        {"role": "system", "content": "You are a friendly Chatbot. Always reply in the language in which the user is writing to you."}
+    ]
+    
     for msg in state["messages"]:
         if isinstance(msg, HumanMessage):
             messages.append({"role": "user", "content": msg.content})