Update app.py

app.py

@@ -45,7 +45,7 @@ def find_most_similar_data(query):
 
     return most_similar
 
-async def respond_with_prefix(message, history, max_tokens=2048, temperature=1.0, top_p=1.0) -> Iterator[str]:
+async def respond_with_prefix(message, history, max_tokens=3648, temperature=1.0, top_p=1.0) -> Iterator[str]:
     global conversation_history
     
     # 대화 히스토리 업데이트
@@ -110,22 +110,23 @@ async def respond_with_prefix(message, history, max_tokens=2048, temperature=1.0
         messages.append({"role": "assistant", "content": f"Related Information: {similar_data}"})
 
     try:
-        response = await asyncio.to_thread(
-            lambda: client.chat.completions.create(
-                model="gpt-4o-mini",
-                messages=messages,
-                response_format={"type": "text"},
-                temperature=temperature,
-                max_tokens=max_tokens,
-                top_p=top_p,
-                frequency_penalty=0,
-                presence_penalty=0,
-                stream=True
-            )
+        response = client.chat.completions.create(
+            model="gpt-4o-mini",
+            messages=[
+                {"role": "system", "content": system_prefix},
+                *conversation_history
+            ],
+            response_format={"type": "text"},
+            temperature=temperature,
+            max_tokens=max_tokens,
+            top_p=top_p,
+            frequency_penalty=0,
+            presence_penalty=0,
+            stream=True
         )
 
         partial_message = ""
-        async for chunk in response:
+        for chunk in response:
             if hasattr(chunk.choices[0].delta, 'content') and chunk.choices[0].delta.content:
                 content = chunk.choices[0].delta.content
                 partial_message += content
@@ -144,19 +145,19 @@ def clear_history():
     conversation_history = []
     return None
 
-demo = gr.ChatInterface(
-    fn=respond_with_prefix,
-    additional_inputs=[
-        gr.Slider(minimum=1, maximum=4096, value=2048, label="Max Tokens"),
-        gr.Slider(minimum=0.1, maximum=2.0, value=1.0, label="Temperature"),
-        gr.Slider(minimum=0.1, maximum=1.0, value=1.0, label="Top-P")
-    ],
-    theme="Nymbo/Nymbo_Theme"
-)
-
-# Clear 버튼 추가
-clear_button = gr.Button("Clear History")
-clear_button.click(fn=clear_history)
+with gr.Blocks(theme="Nymbo/Nymbo_Theme") as demo:
+    chatbot = gr.ChatInterface(
+        fn=respond_with_prefix,
+        additional_inputs=[
+            gr.Slider(minimum=1, maximum=4096, value=2048, label="Max Tokens"),
+            gr.Slider(minimum=0.1, maximum=2.0, value=1.0, label="Temperature"),
+            gr.Slider(minimum=0.1, maximum=1.0, value=1.0, label="Top-P")
+        ],
+    )
+    
+    with gr.Row():
+        clear_button = gr.Button("Clear History")
+        clear_button.click(fn=clear_history, outputs=chatbot.chatbot)
 
 if __name__ == "__main__":
     demo.queue(max_size=4).launch()