[Core][Refactor] Move KV cache planning logic to kv_cache_planning.py (#61183)
Signed-off-by: Dakai An <dakaian108@gmail.com> Signed-off-by: Yifan Qiao <yifanqiao@inferact.ai> Co-authored-by: Yifan Qiao <yifanqiao@inferact.ai> Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
co-authored by
Yifan Qiao
Claude Opus 5.5
parent
c077dc663d
commit
df6c7ad031
@@ -43,7 +43,10 @@ from vllm.v1.attention.backends.mla.rocm_aiter_mla_sparse import (
|
||||
ROCMAiterMLASparseBackend,
|
||||
)
|
||||
from vllm.v1.attention.backends.registry import AttentionBackendEnum
|
||||
from vllm.v1.core.kv_cache_utils import estimate_max_model_len, get_kv_cache_configs
|
||||
from vllm.v1.core.kv_cache_planning import (
|
||||
estimate_max_model_len,
|
||||
get_kv_cache_configs,
|
||||
)
|
||||
from vllm.v1.core.sched.output import CachedRequestData, NewRequestData, SchedulerOutput
|
||||
from vllm.v1.kv_cache_interface import (
|
||||
FullAttentionSpec,
|
||||
|
||||
Reference in New Issue
Block a user