[Refactor] move deep_gemm_wrapper out of quantization (#11784)
This commit is contained in:
@@ -8,6 +8,7 @@ from typing import TYPE_CHECKING, Dict, List, Optional, Sequence
|
||||
|
||||
import torch
|
||||
|
||||
from sglang.srt.layers import deep_gemm_wrapper
|
||||
from sglang.srt.layers.attention.base_attn_backend import AttentionBackend
|
||||
from sglang.srt.layers.communicator import (
|
||||
CommunicateContext,
|
||||
@@ -24,7 +25,6 @@ from sglang.srt.layers.moe.token_dispatcher import (
|
||||
DeepEPDispatcher,
|
||||
MooncakeEPDispatcher,
|
||||
)
|
||||
from sglang.srt.layers.quantization import deep_gemm_wrapper
|
||||
from sglang.srt.managers.schedule_batch import ScheduleBatch
|
||||
from sglang.srt.model_executor.forward_batch_info import (
|
||||
ForwardBatch,
|
||||
|
||||
Reference in New Issue
Block a user