Update app.py

app.py

@@ -1,3 +1,22 @@
-import gradio as gr
+import gradio as gr 
+import huggingface_hub as hf_hub 
+import os 
 
-gr.load("models/google/flan-t5-xxl").launch()
+hf_client = hf_hub.InferenceClient(token = os.environ['HF_TOKEN']) 
+
+
+def interface(question):
+    response = hf_client.text_generation(
+        prompt = question,
+        model = 'google/flan-t5-xxl',
+        max_new_tokens = 1024,
+        stream = False
+    )
+
+    return response
+
+app = gr.Interface(
+    fn = interface,
+    inputs = gr.Textbox(label = 'Input Question'),
+    outputs = gr.Textbox(label = 'Generated Response')
+)