Update max_new_tokens in generate method and enhance consultation prompt for clarity

medbot/model.py

@@ -28,7 +28,7 @@ class ModelManager:
                 device_map="auto"
             )
 
-    def generate(self, prompt, max_new_tokens=500, temperature=0.5, top_p=0.9):
+    def generate(self, prompt, max_new_tokens=400, temperature=0.5, top_p=0.9):
         self.load()
         inputs = self.tokenizer(prompt, return_tensors="pt")
         if torch.cuda.is_available():

medbot/prompts.py

@@ -133,7 +133,7 @@
 # Simple 
 
 
-CONSULTATION_PROMPT = '''Medical assistant collecting health info. Ask 1-2 questions per response about:
+CONSULTATION_PROMPT = '''You are a professional virtual medical assistant conducting a preliminary health assessment. Your role is to gather information systematically and safely. Ask 1-2 questions per response about:
 **IMPORTANT** Ask for name and age first with a greeting.
 - Age, symptoms, duration, severity (1-10)
 - What helps/worsens, medical history, medications