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 2, 18:15 UTC
Fix test_aot_compile for torch 2.12 (#43695)
Signed-off-by: Angela Yi <yiangela7@gmail.com>
Angela Yi committed
May 26, 2026 at 23:12 UTC
0fa3114ae11ff8bfdba3e6fb26ad4c4d093ad25a
1 file changed
+4
-1
tests/compile/test_aot_compile.py
+4
-1
@@ -465,7 +465,10 @@ def test_standalone_compile_correctness():
465
common_args,
466
common_args,
467
env1={"VLLM_USE_STANDALONE_COMPILE": "1"},
468
- env2={"VLLM_USE_STANDALONE_COMPILE": "0"},
468
+ env2={
469
+ "VLLM_USE_STANDALONE_COMPILE": "0",
470
+ "VLLM_USE_MEGA_AOT_ARTIFACT": "0",
471
+ },
472
)
473
474