Update main.py

main.py

@@ -5,6 +5,7 @@ from pydantic import BaseModel
 # Model loading
 llm = AutoModelForCausalLM.from_pretrained(
     "Meta-Llama-3-8B-Instruct.Q4_K_M.gguf",
+    model_type='llama',
     max_new_tokens=1096,
     threads=3,
 )