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 7, 15:15 UTC
[compile] Invalidate cache for cpu flags (#35119)
Signed-off-by: angelayi <yiangela7@gmail.com>
Angela Yi committed
Feb 26, 2026 at 18:54 UTC
c29ee9c32647cc6cc3c51a6bc070267d48b0bcc4
1 file changed
-3
vllm/envs.py
-3
@@ -1752,10 +1752,7 @@ def compile_factors() -> dict[str, object]:
1752
"VLLM_ENABLE_V1_MULTIPROCESSING",
1753
"VLLM_V1_OUTPUT_PROC_CHUNK_SIZE",
1754
"VLLM_CPU_KVCACHE_SPACE",
1755
- "VLLM_CPU_OMP_THREADS_BIND",
1756
- "VLLM_CPU_NUM_OF_RESERVED_CPU",
1755
"VLLM_CPU_MOE_PREPACK",
1758
- "VLLM_CPU_SGL_KERNEL",
1756
"VLLM_TEST_FORCE_LOAD_FORMAT",
1757
"VLLM_ENABLE_CUDA_COMPATIBILITY",
1758
"VLLM_CUDA_COMPATIBILITY_PATH",