first commit

app.py

@@ -2,6 +2,9 @@ import spaces
 import gradio as gr
 from transformers import pipeline
 import torch
+import os
+
+hf_token = os.environ["HF_TOKEN"]
 
 # Load the Gemma 3 pipeline.
 # Gemma 3 is a multimodal model that accepts text and image inputs.
@@ -9,7 +12,8 @@ pipe = pipeline(
     "image-text-to-text",
     model="google/gemma-3-4b-it",
     device="cuda" if torch.cuda.is_available() else "cpu",
-    torch_dtype=torch.bfloat16 if torch.cuda.is_available() else None
+    torch_dtype=torch.bfloat16 if torch.cuda.is_available() else None,
+    use_auth_token=hf_token
 )
 @spaces.GPU
 def generate_response(user_text, user_image, history):