Update app.py

app.py

@@ -16,7 +16,7 @@ streamer = TextIteratorStreamer(tokenizer, timeout=10.0, skip_prompt=True, skip_
 
 @spaces.GPU
 def generate_text(prompt, max_length, top_p, top_k):
-    inputs = tokenizer([prompt], return_tensors="pt")
+    inputs = tokenizer([prompt], return_tensors="pt").to("cuda")
 
     generate_kwargs = dict(
         inputs,