Update app.py

app.py

@@ -15,17 +15,19 @@ ACCESS_TOKEN = os.getenv("HF_TOKEN", "")
 model_id = "bartowski/Mistral-Nemo-Instruct-2407-GGUF"
 filename = "Mistral-Nemo-Instruct-2407-Q6_K_L.gguf"
 
+@spaces.GPU
 model = AutoModelForCausalLM.from_pretrained(
     model_id,
     gguf_file=filename,
-    #torch_dtype=torch.float16,
+    torch_dtype=torch.float16,
     device_map="auto",
-    trust_remote_code=True,
+    trust_remote_code=False,
     token=ACCESS_TOKEN)
+@spaces.GPU
 tokenizer = AutoTokenizer.from_pretrained(
     model_id,
     gguf_file=filename,
-    trust_remote_code=True,
+    trust_remote_code=False,
     token=ACCESS_TOKEN)
 
 tokenizer.use_default_system_prompt = False