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 14, 05:30 UTC
[Bugfix] Fix triton import precommit failure (#25803)
Signed-off-by: Tyler Michael Smith <tlrmchlsmth@gmail.com>
Tyler Michael Smith committed
Sep 27, 2025 at 10:13 UTC
f9df8b4ad77a933659c93cb1b923c8e09d76ea3a
1 file changed
+2
-2
vllm/model_executor/layers/batch_invariant.py
+2
-2
@@ -7,8 +7,8 @@ from collections.abc import Callable
7
from typing import Any, Union
8
9
import torch
10
-import triton
11
-import triton.language as tl
10
+
11
+from vllm.triton_utils import tl, triton
12
13
14
def _matmul_launch_metadata(grid: Callable[..., Any], kernel: Any,