Update app.py

app.py

@@ -15,6 +15,7 @@ llm = Llama(
     n_ctx=2048,
     chat_format="llama-3",
     n_gpu_layers=-1,  # ensure all layers are on GPU
+    flash_attn=True,
 )
 
 # Placeholder responses for when context is empty
@@ -62,7 +63,6 @@ def bot(history):
         temperature=0.7,
         top_p=0.95,
         stream=True,
-        num_threads=1,
     )
     
     for chunk in response: