[Core] Move MRV1 late_interaction_runner.py out of MRV2 subtree (#48014)

Signed-off-by: Nick Hill <[email protected]>
This commit is contained in:
Nick Hill
2026-07-08 18:30:11 +00:00
committed by GitHub
parent 8347c6e6e1
commit a5d19cbb95
3 changed files with 2 additions and 2 deletions
@@ -11,7 +11,7 @@ from vllm.v1.pool.late_interaction import (
build_late_interaction_doc_params,
build_late_interaction_query_params,
)
from vllm.v1.worker.gpu.pool.late_interaction_runner import LateInteractionRunner
from vllm.v1.pool.late_interaction_runner import LateInteractionRunner
def _make_pooling_params(
+1 -1
View File
@@ -174,6 +174,7 @@ from vllm.v1.outputs import (
SamplerOutput,
make_empty_encoder_model_runner_output,
)
from vllm.v1.pool.late_interaction_runner import LateInteractionRunner
from vllm.v1.pool.metadata import PoolingMetadata, PoolingStates
from vllm.v1.sample.logits_processor import LogitsProcessors, build_logitsprocs
from vllm.v1.sample.logits_processor.interface import LogitsProcessor
@@ -207,7 +208,6 @@ from vllm.v1.worker.cp_utils import (
from vllm.v1.worker.dp_utils import coordinate_batch_across_dp
from vllm.v1.worker.ec_connector_model_runner_mixin import ECConnectorModelRunnerMixin
from vllm.v1.worker.gpu.attn_utils import _reshape_attention_kv_cache
from vllm.v1.worker.gpu.pool.late_interaction_runner import LateInteractionRunner
from vllm.v1.worker.gpu_input_batch import CachedRequestState, InputBatch
from vllm.v1.worker.gpu_ubatch_wrapper import UBatchWrapper
from vllm.v1.worker.kv_connector_model_runner_mixin import KVConnectorModelRunnerMixin