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 13, 20:45 UTC
[Model] fix ernie45 reasoning_parser (#27973)
Signed-off-by: wangyafeng <wangyafeng@baidu.com>
CSWYF3634076 committed
Nov 4, 2025 at 15:16 UTC
43a6acfb7de8c7ad839d41bc2109fafe692b77ba
1 file changed
+2
-2
vllm/reasoning/ernie45_reasoning_parser.py
+2
-2
@@ -36,8 +36,8 @@ class Ernie45ReasoningParser(BaseThinkingReasoningParser):
36
"""The token that ends reasoning content."""
37
return "</think>"
38
39
- def __init__(self, tokenizer: PreTrainedTokenizerBase):
40
- super().__init__(tokenizer)
39
+ def __init__(self, tokenizer: PreTrainedTokenizerBase, *args, **kwargs):
40
+ super().__init__(tokenizer, *args, **kwargs)
41
42
if not self.model_tokenizer:
43
raise ValueError(