Update app.py

app.py

@@ -1,5 +1,5 @@
 import gradio as gr
-from transformers import AutoModelForCausalLM, LlamaTokenizerFast
+from transformers import AutoModelForCausalLM, LlamaTokenizer
 import torch
 from huggingface_hub import login
 import re
@@ -10,7 +10,7 @@ login(token=os.getenv("HF_TOKEN"))
 
 # Load the model and tokenizer
 model_name = "ranggafermata/Fermata-v1.2-lightcoder"
-tokenizer = LlamaTokenizerFast.from_pretrained(model_name)
+tokenizer = LlamaTokenizer.from_pretrained(model_name)
 model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32)
 model.eval()