[V0 Deprecation] Remove V0 sampling metadata (#25345)

Signed-off-by: Woosuk Kwon <woosuk@thinkingmachines.ai>
This commit is contained in:
Woosuk Kwon
2025-09-21 10:37:11 -07:00
committed by GitHub
parent c438b2951c
commit 1c3ffdbecc
141 changed files with 172 additions and 583 deletions
+1 -4
View File
@@ -21,7 +21,6 @@ from vllm.attention.layer import check_upstream_fa_availability
from vllm.config import VllmConfig
from vllm.distributed import get_tensor_model_parallel_world_size
from vllm.logger import init_logger
from vllm.model_executor import SamplingMetadata
from vllm.model_executor.layers.linear import (ColumnParallelLinear,
QKVParallelLinear,
RowParallelLinear)
@@ -1556,10 +1555,8 @@ class BaseKeyeModule(nn.Module):
def compute_logits(
self,
hidden_states: torch.Tensor,
sampling_metadata: SamplingMetadata,
) -> Optional[torch.Tensor]:
return self.language_model.compute_logits(hidden_states,
sampling_metadata)
return self.language_model.compute_logits(hidden_states)
def load_weights(self, weights: Iterable[tuple[str,
torch.Tensor]]) -> set[str]: