[Refactor] Use dynamic MM cache in processor (#57674)

Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
This commit is contained in:
Cyrus Leung
2026-09-19 15:38:00 +00:00
committed by GitHub
parent 41c4a3ed4e
commit 751f6807d9
23 changed files with 343 additions and 401 deletions
-5
View File
@@ -25,7 +25,6 @@ from vllm.model_executor.layers.activation import get_act_fn
from vllm.model_executor.layers.linear import ColumnParallelLinear, RowParallelLinear
from vllm.model_executor.layers.quantization import QuantizationConfig
from vllm.multimodal import MULTIMODAL_REGISTRY
from vllm.multimodal.cache import BaseMultiModalProcessorCache
from vllm.multimodal.inputs import (
MultiModalFieldConfig,
MultiModalKwargsItem,
@@ -407,21 +406,17 @@ def _build_llava_or_pixtral_hf_info(
def _build_llava_or_pixtral_hf_processor(
info: _I,
dummy_inputs: BaseDummyInputsBuilder[_I],
*,
cache: BaseMultiModalProcessorCache | None = None,
) -> BaseMultiModalProcessor:
if isinstance(info, PixtralHFProcessingInfo):
return PixtralHFMultiModalProcessor(
info,
dummy_inputs, # type: ignore
cache=cache,
)
if isinstance(info, LlavaProcessingInfo):
return LlavaMultiModalProcessor(
info,
dummy_inputs, # type: ignore
cache=cache,
)
raise NotImplementedError(type(info))