Update app.py

app.py

@@ -5,27 +5,31 @@ client = InferenceClient("thviet79/model-QA-medical")
 
 def respond(
     message,
-    history: list[tuple[str, str]],
+    history: list[dict],  # Use list of dicts instead of tuples
     system_message,
     max_tokens,
     temperature,
     top_p,
 ):
-    # Simplified version for testing
+    # Prepare messages for the model
     messages = [{"role": "system", "content": system_message}]
 
     for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
+        if val["role"] == "user":
+            messages.append({"role": "user", "content": val["content"]})
+        if val["role"] == "assistant":
+            messages.append({"role": "assistant", "content": val["content"]})
 
     messages.append({"role": "user", "content": message})
 
     # Temporary placeholder response for testing
     response = f"Test response for message: {message}"
-    return response
+    
+    # Update history with assistant's response
+    history.append({"role": "user", "content": message})
+    history.append({"role": "assistant", "content": response})
 
+    return response, history
 
 demo = gr.ChatInterface(
     respond,
@@ -41,6 +45,7 @@ demo = gr.ChatInterface(
             label="Top-p (nucleus sampling)",
         ),
     ],
+    type='messages'  # This makes it use the OpenAI-style structure
 )
 
 if __name__ == "__main__":