Update app.py

app.py

@@ -20,7 +20,7 @@ def respond(
     temperature,
     top_p,
 ):
-    llama_model = Llama(r"models/mistral-7b-instruct-v0.2.Q6_K.gguf", n_batch=1024, n_threads=0, n_gpu_layers=64, n_ctx=8192, verbose=False)
+    llama_model = Llama(r"models/mistral-7b-instruct-v0.2.Q6_K.gguf", n_batch=1024, n_threads=0, n_gpu_layers=33, n_ctx=8192, verbose=False)
 
     provider = LlamaCppPythonProvider(llama_model)