Update app.py

app.py

@@ -4,12 +4,14 @@ from llama_cpp import Llama
 try:
     llm = Llama.from_pretrained(
         repo_id="operablepattern/gemma-2b-it-Q",
-        filename="*Q4_K_M.gguf",
+        filename="*Q5_K_M.gguf",
+        max_tokens=32,
         verbose=True
     )
 except:
     llm = Llama(
         model_path="./gemma-2b-it-Q5_K_M.gguf",
+        max_tokens=32,
         verbose=True
     )
 
@@ -21,5 +23,4 @@ def response(message, history):
 gr.ChatInterface(
     fn=response,
     title="Chat with Gemma",
-    theme='sky'
 ).queue().launch()