Update app.py

app.py

@@ -46,17 +46,9 @@ async def chat_completion(request: Request):
         data = await request.json()
         messages = data.get("messages", [])
         
-        # Format messages for Qwen
-        conversation = []
-        for msg in messages:
-            conversation.append({
-                "role": msg["role"],
-                "content": msg["content"]
-            })
-        
         # Convert messages to model input format
         prompt = tokenizer.apply_chat_template(
-            conversation,
+            messages,
             tokenize=False,
             add_generation_prompt=True
         )
@@ -89,15 +81,23 @@ async def chat_completion(request: Request):
 # Gradio interface for testing
 def chat_interface(message, history):
     history = history or []
-    messages = [{"role": "user", "content": message}]
+    messages = []
+    
+    # Convert history to messages format
+    for user_msg, assistant_msg in history:
+        messages.append({"role": "user", "content": user_msg})
+        messages.append({"role": "assistant", "content": assistant_msg})
     
-    # Add history to messages
-    for h in history:
-        messages.insert(0, {"role": "assistant" if i % 2 else "user", "content": h[1 if i % 2 else 0]} 
-                       for i in range(len(h)))
+    # Add current message
+    messages.append({"role": "user", "content": message})
     
-    response = chat_completion(Request({"messages": messages}))
-    return response.choices[0].message.content
+    # Get response
+    response = chat_completion(Request(scope={"type": "http"}, receive=None))
+    if isinstance(response, JSONResponse):
+        response_data = response.body.decode()
+        response_json = json.loads(response_data)
+        return response_json["choices"][0]["message"]["content"]
+    return "Error generating response"
 
 interface = gr.ChatInterface(
     chat_interface,