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 12, 17:15 UTC
Handle triton kernel import exception (#29062)
jinghanhu committed
Nov 22, 2025 at 18:07 UTC
988ee66b0d54ec08a24135f7a947affe69e9dd52
1 file changed
+3
-2
vllm/model_executor/layers/fused_moe/config.py
+3
-2
@@ -28,10 +28,11 @@ logger = init_logger(__name__)
28
if has_triton_kernels():
29
try:
30
from triton_kernels.matmul_ogs import PrecisionConfig
31
- except ImportError:
31
+ except (ImportError, AttributeError) as e:
32
logger.error(
33
"Failed to import Triton kernels. Please make sure your triton "
34
- "version is compatible."
34
+ "version is compatible. Error: %s",
35
+ e,
36
)
37
38