[Bugfix] Register VLLM_BATCH_INVARIANT in envs.py to fix spurious unknown env var warning (#35007)

Signed-off-by: Ranran <1012869439@qq.com>
Signed-off-by: Ranran <hzz5361@psu.edu>
Signed-off-by: ran <hzz5361@psu.edu>
Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>
This commit is contained in:
Ranran
2026-03-23 18:31:14 -04:00
committed by GitHub
co-authored by Wentao Ye
parent e85f8f0932
commit dc6908ac6a
30 changed files with 70 additions and 130 deletions
@@ -6,7 +6,6 @@ from collections.abc import Sequence
import torch
import vllm.envs as envs
from vllm.model_executor.layers.batch_invariant import vllm_is_batch_invariant
from vllm.model_executor.layers.quantization.utils.fp8_utils import (
process_fp8_weight_block_strategy,
)
@@ -42,7 +41,7 @@ class MarlinFP8ScaledMMLinearKernel(FP8ScaledMMLinearKernel):
# Check if platform supports FP8 Marlin
if not is_fp8_marlin_supported():
return False, "FP8 Marlin requires compute capability 7.5 or higher"
if vllm_is_batch_invariant():
if envs.VLLM_BATCH_INVARIANT:
return False, "FP8 Marlin not supported for batch invariant execution."
if (
compute_capability is not None