Update app.py

app.py

@@ -3,12 +3,12 @@ from transformers import T5Tokenizer, T5ForConditionalGeneration
 from transformers import AutoModel
 
 # xl size run out of memory on 16GB VM
-# google/flan-t5-large
-model_name = 'jncraton/fastchat-t5-3b-v1.0-ct2-int8'
+model_name =  'google/flan-t5-large'
+#model_name = 'jncraton/fastchat-t5-3b-v1.0-ct2-int8'
 
 # Load model directly
 #from transformers import AutoModel
-model = AutoModel.from_pretrained("jncraton/fastchat-t5-3b-v1.0-ct2-int8")
+model = AutoModel.from_pretrained(model_name)
 
 tokenizer = T5Tokenizer.from_pretrained(model_name)
 #model = T5ForConditionalGeneration.from_pretrained(model_name)