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
[Misc] Modify CacheConfig import (#23459)
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
Jee Jee Li committed
Aug 23, 2025 at 14:05 UTC
65197a5fb37ef4d8b93e0b99ecc8b902fe948e97
1 file changed
+1
-1
vllm/attention/layers/encoder_only_attention.py
+1
-1
@@ -5,13 +5,13 @@ from copy import copy
5
from typing import Optional
6
7
import torch
8
-from transformers import CacheConfig
8
9
from vllm import envs
10
from vllm.attention.backends.abstract import (AttentionBackend,
11
AttentionMetadata, AttentionType)
12
from vllm.attention.layer import Attention
13
from vllm.attention.selector import get_attn_backend
14
+from vllm.config import CacheConfig
15
from vllm.v1.attention.backends.utils import (CommonAttentionMetadata,
16
subclass_attention_backend)
17