Update app.py

app.py

@@ -10,7 +10,6 @@ MODEL_PATH = "astegaras/merged_kaggle"   # HF repo containing your .gguf
 # llama_cpp automatically downloads from HF Hub if you provide the repo
 llm = Llama.from_pretrained(
     repo_id=MODEL_PATH,
-    filename="llama-3.2-3b-instruct.Q4_K_M.gguf",
     n_threads=8,
     n_ctx=4096,
     verbose=False,