[Fix] epd: Accept tokens-in requests on scheduler-side transfer backends (#41376)
Signed-off-by: Krishnan Prashanth <kprashanth@nvidia.com> Co-authored-by: Shangming Cai <csmthu@gmail.com> Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
co-authored by
Shangming Cai
Claude Opus 5.5
parent
82f7249ce6
commit
f68e82e7bc
@@ -787,9 +787,12 @@ class WaitingMMRequestStatus(IntEnum):
|
||||
|
||||
def _select_mm_processor_prompt(recv_req, mm_processor):
|
||||
"""Mirror tokenizer-side prompt selection for scheduler-side EPD rebuilds."""
|
||||
if mm_processor.prefer_tokenized_input and recv_req.input_ids is not None:
|
||||
return list(recv_req.input_ids)
|
||||
return recv_req.input_text or recv_req.input_ids
|
||||
# The scheduler holds input_ids as array("q"); processors accept str or
|
||||
# list[int], not array.
|
||||
input_ids = None if recv_req.input_ids is None else list(recv_req.input_ids)
|
||||
if mm_processor.prefer_tokenized_input and input_ids is not None:
|
||||
return input_ids
|
||||
return recv_req.input_text or input_ids
|
||||
|
||||
|
||||
def create_part_req_id(original_req_id: str, part_idx: int) -> str:
|
||||
|
||||
Reference in New Issue
Block a user