Update app.py

app.py

@@ -10,7 +10,7 @@ hf_client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
 local_model_name = "codewithdark/latent-recurrent-depth-lm"
 tokenizer = AutoTokenizer.from_pretrained(local_model_name)
 config = AutoConfig.from_pretrained(local_model_name)
-model = AutoModel.from_pretrained(local_model_name, config=config)
+model = AutoModel.from_pretrained(local_model_name, config=config, trust_remote_code=True)
 device = "cuda" if torch.cuda.is_available() else "cpu"
 model.to(device)