diff --git a/vllm/transformers_utils/tokenizer.py b/vllm/transformers_utils/tokenizer.py index 24ddd35abea6..6a31a4198069 100644 --- a/vllm/transformers_utils/tokenizer.py +++ b/vllm/transformers_utils/tokenizer.py @@ -50,11 +50,12 @@ def decode_tokens( `skip_special_tokens=None` means to use the backend's default settings. """ + decode_method = getattr(tokenizer, "_decode", tokenizer.decode) if skip_special_tokens is not None: - return tokenizer.decode(token_ids, - skip_special_tokens=skip_special_tokens) + return decode_method(token_ids, + skip_special_tokens=skip_special_tokens) - return tokenizer.decode(token_ids) + return decode_method(token_ids) def encode_tokens(