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 5, 02:15 UTC
[Bugfix] Accept **kwargs in MiniMaxM2Parser.__init__() (#39861)
Signed-off-by: Tim Messerschmidt <timmesserschmidt@gmail.com>
Tim Messerschmidt committed
Apr 16, 2026 at 09:18 UTC
8d7c9628337aebe0f78239de0b385e82c631abc6
1 file changed
+9
-3
vllm/parser/minimax_m2_parser.py
+9
-3
@@ -43,11 +43,17 @@ class MiniMaxM2Parser(DelegatingParser):
43
reasoning_parser_cls = MiniMaxM2ReasoningParser
44
tool_parser_cls = MinimaxM2ToolParser
45
46
- def __init__(self, tokenizer: TokenizerLike, tools: list[Tool] | None = None):
47
- super().__init__(tokenizer)
46
+ def __init__(
47
+ self,
48
+ tokenizer: TokenizerLike,
49
+ tools: list[Tool] | None = None,
50
+ *args,
51
+ **kwargs,
52
+ ):
53
+ super().__init__(tokenizer, *args, **kwargs)
54
55
# Initialize the underlying parsers
50
- self._reasoning_parser = MiniMaxM2ReasoningParser(tokenizer)
56
+ self._reasoning_parser = MiniMaxM2ReasoningParser(tokenizer, *args, **kwargs)
57
self._tool_parser = MinimaxM2ToolParser(tokenizer, tools)
58
59
logger.debug(