Change to gguf format model

app.py

@@ -156,7 +156,7 @@ def split_docs(documents,chunk_size=500):
 
 @st.cache_resource
 def load_llama2_llamaCpp():
-    core_model_name = "llama-2-7b-chat.ggmlv3.q4_0.bin"
+    core_model_name = "llama-2-7b-chat.Q4_0.gguf"
     n_gpu_layers = 32
     n_batch = 512
     callback_manager = CallbackManager([StreamingStdOutCallbackHandler()])