Update app.py

app.py

@@ -2,10 +2,10 @@ import gradio as gr
 from huggingface_hub import InferenceClient
 import os
 
-# 🔹 Hardcoded token for internal use (or use os.environ for safety)
-HF_TOKEN = os.getenv('telemedpro')
+# 🔹 Load HF token from secret
+HF_TOKEN = os.environ.get('telemedpro')  # store your HF token in Space Secrets
 
-# Persona system message
+# 🔹 Fixed persona system message
 PERSONA_MESSAGE = (
     "You are Dr. Alex, a highly knowledgeable yet empathetic doctor. "
     "You always provide clear, safe, and well-structured medical advice in simple language. "
@@ -13,36 +13,32 @@ PERSONA_MESSAGE = (
     "You behave politely, patiently, and with care, like a trusted family doctor."
 )
 
-# Create HF InferenceClient
-client = InferenceClient(
-    model="m42-health/Llama3-Med42-70B",
-    token=HF_TOKEN
-)
+# 🔹 Create InferenceClient once
+client = InferenceClient(token=HF_TOKEN, model="m42-health/Llama3-Med42-70B")
 
-# Respond function with safe error handling
 def respond(message, history, max_tokens, temperature, top_p):
     try:
-        # Initialize messages with persona
         messages = [{"role": "system", "content": PERSONA_MESSAGE}]
 
-        # Add conversation history safely
-        for h in history:
-            user_msg = h[0] if h[0] is not None else ""
-            ai_msg = h[1] if h[1] is not None else ""
-            messages.append({"role": "user", "content": user_msg})
-            messages.append({"role": "assistant", "content": ai_msg})
+        # Add previous conversation
+        if history:
+            for h in history:
+                user_msg = h[0] if h[0] else ""
+                ai_msg = h[1] if h[1] else ""
+                messages.append({"role": "user", "content": user_msg})
+                messages.append({"role": "assistant", "content": ai_msg})
 
         # Add current user message
         messages.append({"role": "user", "content": message})
 
-        # Stream AI response token by token
         response = ""
+        # Stream model output token by token
         for msg in client.chat_completion(
             messages,
             max_tokens=max_tokens,
             stream=True,
             temperature=temperature,
-            top_p=top_p
+            top_p=top_p,
         ):
             if msg.choices and msg.choices[0].delta.content:
                 token = msg.choices[0].delta.content
@@ -50,7 +46,6 @@ def respond(message, history, max_tokens, temperature, top_p):
                 yield response
 
     except Exception as e:
-        # Return error as a string so Space never crashes
         yield f"⚠️ Space error: {e}"
 
 # 🔹 Gradio Chat Interface
@@ -66,10 +61,9 @@ chatbot = gr.ChatInterface(
 
 # 🔹 Layout
 with gr.Blocks() as demo:
-    with gr.Row():
-        gr.Markdown("## 🩺 AI Health Mentor — Dr. Alex")
+    gr.Markdown("## 🩺 AI Health Mentor — Dr. Alex")
     chatbot.render()
 
-# 🔹 Launch the Space safely
+# 🔹 Launch Space
 if __name__ == "__main__":
     demo.launch(show_error=True)