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, 18:15 UTC
add kimi reasoning parser (#28128)
Signed-off-by: wangzhengtao <wangzhengtao@msh.team> Co-authored-by: wangzhengtao <wangzhengtao@msh.team>
bigmoyan committed
Nov 5, 2025 at 21:48 UTC
0606bea2b63f6b5d49e5d04b4561f2e606cc7c45
1 file changed
+4
vllm/reasoning/__init__.py
+4
@@ -48,6 +48,10 @@ _REASONING_PARSERS_TO_REGISTER = {
48
"hunyuan_a13b_reasoning_parser",
49
"HunyuanA13BReasoningParser",
50
),
51
+ "kimi_k2": (
52
+ "deepseek_r1_reasoning_parser",
53
+ "DeepSeekR1ReasoningParser",
54
+ ),
55
"minimax_m2": (
56
"minimax_m2_reasoning_parser",
57
"MiniMaxM2ReasoningParser",