[Core] Move MRV1 `late_interaction_runner.py` out of MRV2 subtree (#48014)
Signed-off-by: Nick Hill <nickhill123@gmail.com>
Nick Hill committed
Jul 8, 2026 at 19:30 UTC
a5d19cbb95872c4b426c06735733568542fa33db
3 files changed
+2
-2
tests/v1/worker/test_late_interaction_runner.py
+1
-1
@@ -11,7 +11,7 @@ from vllm.v1.pool.late_interaction import (
11
build_late_interaction_doc_params,
12
build_late_interaction_query_params,
13
)
14
-from vllm.v1.worker.gpu.pool.late_interaction_runner import LateInteractionRunner
14
+from vllm.v1.pool.late_interaction_runner import LateInteractionRunner
15
16
17
def _make_pooling_params(
vllm/v1/pool/late_interaction_runner.py
renamed
vllm/v1/worker/gpu_model_runner.py
+1
-1
@@ -174,6 +174,7 @@ from vllm.v1.outputs import (
174
SamplerOutput,
175
make_empty_encoder_model_runner_output,
176
)
177
+from vllm.v1.pool.late_interaction_runner import LateInteractionRunner
178
from vllm.v1.pool.metadata import PoolingMetadata, PoolingStates
179
from vllm.v1.sample.logits_processor import LogitsProcessors, build_logitsprocs
180
from vllm.v1.sample.logits_processor.interface import LogitsProcessor
@@ -207,7 +208,6 @@ from vllm.v1.worker.cp_utils import (
208
from vllm.v1.worker.dp_utils import coordinate_batch_across_dp
209
from vllm.v1.worker.ec_connector_model_runner_mixin import ECConnectorModelRunnerMixin
210
from vllm.v1.worker.gpu.attn_utils import _reshape_attention_kv_cache
210
-from vllm.v1.worker.gpu.pool.late_interaction_runner import LateInteractionRunner
211
from vllm.v1.worker.gpu_input_batch import CachedRequestState, InputBatch
212
from vllm.v1.worker.gpu_ubatch_wrapper import UBatchWrapper
213
from vllm.v1.worker.kv_connector_model_runner_mixin import KVConnectorModelRunnerMixin