Update app.py

app.py

@@ -21,8 +21,8 @@ def Sentence_Commpletion(model_name, input):
         outputs = model.generate(inputss.input_ids, max_length=30, num_return_sequences=1)
     elif model_name == "Gemma":
         tokenizer, model = g_tokenizer, g_model
-        input_ids = tokenizer(input, return_tensors="pt").to("cuda")
-        outputs = model.generate(**input_ids, max_new_tokens=32).to("cuda")
+        input_ids = tokenizer(input, return_tensors="pt")
+        outputs = model.generate(**input_ids, max_new_tokens=32)
     return tokenizer.decode(outputs[0])
 
 
@@ -34,4 +34,4 @@ inputs=[gr.Radio(["Bloom", "Gemma"], label="Choose model"),
 outputs="text",
 title="Bloom vs Gemma Sentence completion",)
 
-interface.launch(debug = True)
+interface.launch(share = True)