Update app.py

app.py

@@ -1,3 +1,4 @@
+import os
 from transformers import MllamaForConditionalGeneration, AutoProcessor, TextIteratorStreamer
 from PIL import Image
 import requests
@@ -7,10 +8,25 @@ import gradio as gr
 from gradio import FileData
 import time
 import spaces
+
+# Hugging Face 토큰을 환경 변수에서 가져옴
+hf_token = os.getenv("HF_TOKEN")
+if not hf_token:
+    raise ValueError("HF_TOKEN environment variable not found")
+
 ckpt = "meta-llama/Llama-3.2-11B-Vision-Instruct"
-model = MllamaForConditionalGeneration.from_pretrained(ckpt,
-    torch_dtype=torch.bfloat16).to("cuda")
-processor = AutoProcessor.from_pretrained(ckpt)
+# 토큰을 사용하여 모델과 프로세서 로드
+model = MllamaForConditionalGeneration.from_pretrained(
+    ckpt,
+    torch_dtype=torch.bfloat16,
+    token=hf_token  # 토큰 추가
+).to("cuda")
+
+processor = AutoProcessor.from_pretrained(
+    ckpt,
+    token=hf_token  # 토큰 추가
+)
+
 
 
 @spaces.GPU