Fix CI

vllm/entrypoints/openai/serving_engine.py

@@ -179,18 +179,17 @@ def _validate_prompt_and_tokenize(
  "Only one of prompt or prompt_ids should be provided.")
 
  if prompt_ids is None:
- # The special tokens (BOS) have already been included by the chat template at this point,
+ # The special tokens (BOS) have already been included by
+ # the chat template at this point,
  # so we don't add them again.
  # See how it's done in huggingface/transformers:
  # https://github.com/huggingface/transformers/blob/5962d62bac850cd01ee830ffba880469338c96fd/src/transformers/tokenization_utils_base.py#L1820-L1829
- tokenizer_kwargs = {"add_special_tokens": False}
+ tokenizer_kwargs: Dict[str, Any] = {"add_special_tokens": False}
  if truncate_prompt_tokens is not None:
- tokenizer_kwargs.update(
- {
- "truncation": True,
- "max_length": truncate_prompt_tokens,
- }
- )
+ tokenizer_kwargs.update({
+ "truncation": True,
+ "max_length": truncate_prompt_tokens,
+ })
  input_ids = self.tokenizer(prompt, **tokenizer_kwargs).input_ids
  elif truncate_prompt_tokens is not None:
  input_ids = prompt_ids[-truncate_prompt_tokens:]