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 8, 14:13 UTC
[compile] Enable AOT compile with 2.10 in trunk. (#34155)
Signed-off-by: Zhengxu Chen <zhxchen17@meta.com>
Zhengxu Chen committed
Feb 10, 2026 at 10:24 UTC
82e11973cc07909de895a1309ce0f6a2144c576a
1 file changed
+1
-1
vllm/envs.py
+1
-1
@@ -271,7 +271,7 @@ def use_aot_compile() -> bool:
271
272
default_value = (
273
"1"
274
- if is_torch_equal_or_newer("2.11.0.dev") and not disable_compile_cache()
274
+ if is_torch_equal_or_newer("2.10.0") and not disable_compile_cache()
275
else "0"
276
)
277