Update app.py

app.py

@@ -14,10 +14,9 @@ title = f"{model_name}"
 description = f"Chat with <a href=\"https://huggingface.co/{model_id}\">{model_name}</a> in GGUF format ({quant})!"
 
 # Initialize the LLM
-llm = Llama(model_path="model.gguf",
-            n_ctx=32768,
-            n_threads=2,
-            chat_format=chat_template)
+#llm = Llama(model_path="model.gguf", n_ctx=32768, n_threads=2, chat_format=chat_template)
+
+llm = Llama(model_path="model.gguf", n_ctx=32768, n_threads=4, n_gpu_layers=12, chat_format=chat_template)
 
 # Function for streaming chat completions
 def chat_stream_completion(message, history, system_prompt):