biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[Misc] Use config definitions from Transformers library (#21913)

Browse Source 
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
This commit is contained in:
Cyrus Leung
2025-08-09 14:10:51 +08:00
committed by GitHub
parent 7ad7adb67f
commit 65552b476b
11 changed files with 54 additions and 51 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
vllm/model_executor/models/qwen2_moe.py
View File

@@ -30,7 +30,7 @@ from typing import Any, Optional, Union
import torch
import torch.nn.functional as F
from torch import nn
from transformers import PretrainedConfig
from transformers import Qwen2MoeConfig
from vllm.attention import Attention
from vllm.compilation.decorators import support_torch_compile
@@ -98,7 +98,7 @@ class Qwen2MoeSparseMoeBlock(nn.Module):
def __init__(
self,
config: PretrainedConfig,
config: Qwen2MoeConfig,
quant_config: Optional[QuantizationConfig] = None,
prefix: str = "",
):
@@ -256,7 +256,7 @@ class Qwen2MoeDecoderLayer(nn.Module):
def __init__(
self,
config: PretrainedConfig,
config: Qwen2MoeConfig,
cache_config: Optional[CacheConfig] = None,
quant_config: Optional[QuantizationConfig] = None,
prefix: str = "",