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
[MISC] fix import violations for re and triton modules (#26654)
Signed-off-by: Sungjae Lee <33976427+llsj14@users.noreply.github.com> Co-authored-by: Mengqing Cao <cmq0113@163.com>
Sungjae Lee committed
Oct 16, 2025 at 19:38 UTC
00417f4e44378bc5c5c5c6698b4d8a86017198d3
1 file changed
+2
-2
vllm/model_executor/layers/quantization/qutlass_utils.py
+2
-2
@@ -14,10 +14,10 @@
14
from typing import Literal
15
16
import torch
17
-import triton
18
-import triton.language as tl
17
from torch.library import wrap_triton
18
19
+from vllm.triton_utils import tl, triton
20
+
21
22
@triton.jit
23
def triton_scale_swizzle(