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 2, 17:01 UTC
[Quantization] Fix Humming RoutedExperts import (#43540)
Signed-off-by: Minh Vu <vuhoangminh97@gmail.com> Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>
Minh Vu committed
May 27, 2026 at 19:51 UTC
206b72c982db29eb10e475589cbf3fa8d6a37071
1 file changed
+1
-1
vllm/model_executor/layers/quantization/utils/humming_utils.py
+1
-1
@@ -8,11 +8,11 @@ from humming.layer import HummingInputSchema, HummingMethod
8
from humming.schema import BaseWeightSchema
9
10
from vllm import envs
11
+from vllm.model_executor.layers.fused_moe import RoutedExperts
12
from vllm.model_executor.layers.fused_moe.config import (
13
FusedMoEQuantConfig,
14
FusedMoEQuantDesc,
15
)
15
-from vllm.model_executor.layers.fused_moe.routed_experts import RoutedExperts
16
from vllm.model_executor.layers.linear import LinearBase
17
from vllm.model_executor.layers.quantization.utils.quant_utils import GroupShape
18