A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 10, 11:30 UTC
[Bugfix]: Fix Step3ReasoningParser missing is_reasoning_end_streaming (#31969)
Signed-off-by: chaunceyjiang <chaunceyjiang@gmail.com>
Chauncey committed
Jan 8, 2026 at 23:28 UTC
eaba8ece7700cbeba380530748847f1f5a672733
1 file changed
+6
vllm/reasoning/step3_reasoning_parser.py
+6
@@ -100,6 +100,12 @@ class Step3ReasoningParser(ReasoningParser):
100
def is_reasoning_end(self, input_ids: list[int]) -> bool:
101
return self.think_end_token_id in input_ids
102
103
+ def is_reasoning_end_streaming(
104
+ self, input_ids: list[int], delta_ids: list[int]
105
+ ) -> bool:
106
+ end_token_id = self.think_end_token_id
107
+ return end_token_id in delta_ids
108
+
109
def extract_content_ids(self, input_ids: list[int]) -> list[int]:
110
if self.think_end_token_id not in input_ids[:-1]:
111
return []