[V0 Deprecation] Remove V0 core (#25321)

Signed-off-by: Woosuk Kwon <woosuk@thinkingmachines.ai>
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
This commit is contained in:
Woosuk Kwon
2025-09-20 19:58:26 -07:00
committed by GitHub
parent 72dd1595b4
commit c99db8c8dd
29 changed files with 24 additions and 9020 deletions
+2 -7
View File
@@ -35,9 +35,6 @@ PAD_SLOT_ID = -1
# if we have at least this many elements. Could be tuned further.
_COMPUTE_SLOT_MAPPING_NUMPY_NUMEL = 256
if TYPE_CHECKING:
from vllm.worker.model_runner import ModelInputForGPUBuilder
def is_block_tables_empty(block_tables: Union[None, Dict]):
"""
@@ -129,7 +126,7 @@ class CommonMetadataBuilder(AttentionMetadataBuilder[TAttentionMetadata]):
_metadata_cls: Type[TAttentionMetadata]
def __init__(self, input_builder: "ModelInputForGPUBuilder"):
def __init__(self, input_builder):
self.input_builder = input_builder
self.runner = input_builder.runner
@@ -149,9 +146,7 @@ class CommonMetadataBuilder(AttentionMetadataBuilder[TAttentionMetadata]):
self.num_prefill_tokens = 0
self.num_decode_tokens = 0
def _add_seq_group(
self, inter_data: "ModelInputForGPUBuilder.InterDataForSeqGroup",
chunked_prefill_enabled: bool):
def _add_seq_group(self, inter_data, chunked_prefill_enabled: bool):
is_prompt = inter_data.is_prompt
block_tables = inter_data.block_tables