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 14, 05:30 UTC
[compile] Fix inductor partition config (#26645)
Signed-off-by: angelayi <yiangela7@gmail.com>
Angela Yi committed
Oct 11, 2025 at 14:03 UTC
01653a917b140535cb6768c995a4bd5ea3253120
1 file changed
+1
-3
vllm/config/compilation.py
+1
-3
@@ -709,9 +709,7 @@ class CompilationConfig:
709
return self.level == CompilationLevel.PIECEWISE
710
711
# Inductor partition case
712
- return (
713
- self.level > CompilationLevel.NO_COMPILATION and self.backend == "inductor"
714
- )
712
+ return self.level > CompilationLevel.NO_COMPILATION and self.use_inductor
713
714
def custom_op_log_check(self):
715
"""