Upload app.py

app.py

@@ -81,7 +81,7 @@ async def predict(message, history):
     async with aiohttp.ClientSession() as session:
         async with session.post(url, headers=headers, json=body) as response:
             
-            buffer = ""  # A buffer to hold incomplete lines of SSE data
+            buffer = ""  # A buffer to hold incomplete lines of data
             async for chunk in response.content.iter_any():
                 buffer += chunk.decode()
                 while "\n" in buffer:  # Process as long as there are complete lines in the buffer
@@ -101,8 +101,8 @@ async def predict(message, history):
                                     full_response += current_text
                                     yield full_response
                                     await asyncio.sleep(0.01)
-                                except Exception as e:
-                                    print(e)
+                                except Exception:
+                                    pass
     
     final_metadata_block = ""
 
@@ -116,11 +116,10 @@ async def predict(message, history):
 
     yield f"{full_response}\n\n{final_metadata_block}"
 
-# Setting up the Gradio chat interface.
 gr.ChatInterface(
     predict,
-    title="Web Search with LLM !",
-    description="Ask any question, and I will try to answer it using web search !",
+    title="Web Search with LLM",
+    description="Ask any question, and I will try to answer it using web search",
     retry_btn=None,
     undo_btn=None,
     examples=[
@@ -129,4 +128,4 @@ gr.ChatInterface(
         'What is the capital of France?',
         'Why does Brazil has a high tax rate?'
     ]
-).launch()  # Launching the web interface.
+).launch()