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

[1/N][Cleanup] Standardize on use of is_quantized_kv_cache (#38659)

Browse Source 
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
This commit is contained in:
Matthew Bonanni
2026-04-01 00:08:01 -04:00
committed by GitHub
parent 7b01d97a22
commit 116f4be405
28 changed files with 90 additions and 75 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
vllm/model_executor/models/extract_hidden_states.py
View File

@@ -23,14 +23,13 @@ from vllm.model_executor.layers.attention.kv_transfer_utils import (
)
from vllm.model_executor.layers.attention_layer_base import AttentionLayerBase
from vllm.model_executor.models.utils import maybe_prefix
from vllm.utils.torch_utils import kv_cache_dtype_str_to_dtype
from vllm.utils.torch_utils import is_quantized_kv_cache, kv_cache_dtype_str_to_dtype
from vllm.v1.attention.backend import (
AttentionBackend,
AttentionImpl,
AttentionMetadataBuilder,
AttentionType,
CommonAttentionMetadata,
is_quantized_kv_cache,
)
from vllm.v1.kv_cache_interface import (
AttentionSpec,