Change device to CPU.

app.py

@@ -14,11 +14,13 @@ import tiktoken
 compressors = {
     "xlm-roberta-large": PromptCompressor(
         model_name="microsoft/llmlingua-2-xlm-roberta-large-meetingbank",
-        use_llmlingua2=True
+        use_llmlingua2=True,
+        device="cpu"
     ),
     "mbert-base": PromptCompressor(
         model_name="microsoft/llmlingua-2-bert-base-multilingual-cased-meetingbank",
-        use_llmlingua2=True
+        use_llmlingua2=True,
+        device="cpu"
     )
 }
 tokenizer = tiktoken.encoding_for_model("gpt-4")