[Bugfix] Register VLLM_BATCH_INVARIANT in envs.py to fix spurious unknown env var warning

vllm/envs.py

@@ -74,6 +74,7 @@
     VLLM_TARGET_DEVICE: str = "cuda"
     VLLM_MAIN_CUDA_VERSION: str = "12.9"
     VLLM_FLOAT32_MATMUL_PRECISION: Literal["highest", "high", "medium"] = "highest"
+    VLLM_BATCH_INVARIANT: bool = False
     MAX_JOBS: str | None = None
     NVCC_THREADS: str | None = None
     VLLM_USE_PRECOMPILED: bool = False
@@ -484,6 +485,9 @@ def _get_or_set_default() -> str:
         ["highest", "high", "medium"],
         case_sensitive=False,
     ),
+    # Enable batch-invariant mode: deterministic results regardless of
+    # batch composition. Requires NVIDIA GPU with compute capability >= 9.0.
+    "VLLM_BATCH_INVARIANT": lambda: bool(int(os.getenv("VLLM_BATCH_INVARIANT", "0"))),
     # Maximum number of compilation jobs to run in parallel.
     # By default this is the number of CPUs
     "MAX_JOBS": lambda: os.getenv("MAX_JOBS", None),