Update app.py

app.py

@@ -18,9 +18,9 @@ def configure_llama_model():
     llm = LlamaCPP(
         model_url=model_url,
         temperature=0.3,
-        max_new_tokens=256,
-        context_window=2000,
-        model_kwargs={"n_gpu_layers": -1},
+        max_new_tokens=68,
+        context_window=1000,
+        model_kwargs={"n_gpu_layers": 1},
         messages_to_prompt=messages_to_prompt,
         completion_to_prompt=completion_to_prompt,
         verbose=True,