[Structured Output] Remove redundant check for grammar_bitmask (#1459)
### What this PR does / why we need it? Remove redundant check since we have check this at https://github.com/vllm-project/vllm-ascend/blob/main/vllm_ascend/worker/model_runner_v1.py#L1450. Signed-off-by: shen-shanshan <467638484@qq.com>
This commit is contained in:
@@ -1273,8 +1273,6 @@ class NPUModelRunner(LoRAModelRunnerMixin):
|
||||
# Serialization of np.ndarray is much more efficient than a tensor,
|
||||
# so we receive it in that format.
|
||||
grammar_bitmask = scheduler_output.grammar_bitmask
|
||||
if grammar_bitmask is None:
|
||||
return
|
||||
|
||||
# We receive the structured output bitmask from the scheduler, but the
|
||||
# indices of the requests in the batch may not match the indices of
|
||||
|
||||
Reference in New Issue
Block a user