Change to smaller llama3

app.py

@@ -4,7 +4,7 @@ from llama_cpp import Llama
 model = "SanctumAI/Meta-Llama-3-8B-Instruct-GGUF"
 llm = Llama.from_pretrained(
     repo_id=model,
-    filename="meta-llama-3-8b-instruct.Q8_0.gguf",
+    filename="meta-llama-3-8b-instruct.Q4_K_S.gguf",
     verbose=True,
     use_mmap=False,
     use_mlock=True,