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 for the import error from #24588 (#25481)
Signed-off-by: Gregory Shtrasberg <Gregory.Shtrasberg@amd.com>
Gregory Shtrasberg committed
Sep 23, 2025 at 17:31 UTC
690f948e4a440baa85a7327c7e24f297d25b9bb2
1 file changed
+1
-1
vllm/model_executor/layers/fused_moe/gpt_oss_triton_kernels_moe.py
+1
-1
@@ -23,7 +23,7 @@ if has_triton_kernels():
23
from triton_kernels.routing import (RoutingData, routing,
24
routing_from_bitmatrix)
25
from triton_kernels.tensor import Bitmatrix
26
- except (ModuleNotFoundError, AttributeError) as e:
26
+ except (AttributeError, ImportError) as e:
27
logger.error(
28
"Failed to import Triton kernels. Please make sure your triton "
29
"version is compatible. Error: %s", e)