[Misc] Reorganize inputs (#35182)

Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>

vllm/model_executor/models/ernie45_vl.py

@@ -40,6 +40,7 @@ from vllm.config import VllmConfig
 from vllm.config.multimodal import BaseDummyOptions, VideoDummyOptions
 from vllm.distributed import parallel_state
 from vllm.distributed import utils as dist_utils
+from vllm.inputs import MultiModalDataDict
 from vllm.logger import init_logger
 from vllm.model_executor.layers.activation import QuickGELU
 from vllm.model_executor.layers.attention import (
@@ -58,7 +59,6 @@ from vllm.model_executor.layers.rotary_embedding.common import (
 from vllm.model_executor.model_loader.weight_utils import default_weight_loader
 from vllm.multimodal import MULTIMODAL_REGISTRY
 from vllm.multimodal.inputs import (
-    MultiModalDataDict,
     MultiModalFeatureSpec,
     MultiModalFieldConfig,
     MultiModalKwargsItems,

vllm/model_executor/models/fireredasr2.py

@@ -15,7 +15,7 @@ from transformers import (
 
 from vllm.config import ModelConfig, SpeechToTextConfig, VllmConfig
 from vllm.config.multimodal import BaseDummyOptions
-from vllm.inputs.data import PromptType
+from vllm.inputs import MultiModalDataDict, PromptType
 from vllm.logger import init_logger
 from vllm.model_executor.layers.activation import _ACTIVATION_REGISTRY
 from vllm.model_executor.layers.linear import (
@@ -27,7 +27,6 @@ from vllm.model_executor.models.whisper_utils import (
 )
 from vllm.multimodal import MULTIMODAL_REGISTRY
 from vllm.multimodal.inputs import (
-    MultiModalDataDict,
     MultiModalFieldConfig,
     MultiModalKwargsItems,
 )

vllm/model_executor/models/funasr.py

@@ -17,7 +17,7 @@ from transformers import (
 from vllm.config import ModelConfig, SpeechToTextConfig, VllmConfig
 from vllm.config.multimodal import BaseDummyOptions
 from vllm.distributed import get_tensor_model_parallel_world_size
-from vllm.inputs.data import PromptType
+from vllm.inputs import MultiModalDataDict, PromptType
 from vllm.logger import init_logger
 from vllm.model_executor.layers.activation import _ACTIVATION_REGISTRY
 from vllm.model_executor.layers.attention.mm_encoder_attention import (
@@ -37,7 +37,6 @@ from vllm.model_executor.models.whisper_utils import (
 )
 from vllm.multimodal import MULTIMODAL_REGISTRY
 from vllm.multimodal.inputs import (
-    MultiModalDataDict,
     MultiModalFieldConfig,
     MultiModalKwargsItems,
 )

vllm/model_executor/models/funaudiochat.py

@@ -27,12 +27,12 @@ from transformers.modeling_outputs import BaseModelOutput
 
 from vllm.config import VllmConfig
 from vllm.config.multimodal import BaseDummyOptions
+from vllm.inputs import MultiModalDataDict
 from vllm.model_executor.layers.attention.mm_encoder_attention import MMEncoderAttention
 from vllm.model_executor.layers.linear import QKVParallelLinear, RowParallelLinear
 from vllm.model_executor.model_loader.weight_utils import default_weight_loader
 from vllm.multimodal import MULTIMODAL_REGISTRY
 from vllm.multimodal.inputs import (
-    MultiModalDataDict,
     MultiModalFieldConfig,
     MultiModalKwargsItems,
 )

vllm/model_executor/models/fuyu.py

@@ -28,11 +28,11 @@ from transformers import BatchFeature, FuyuConfig, FuyuImageProcessor, FuyuProce
 
 from vllm.config import VllmConfig
 from vllm.config.multimodal import BaseDummyOptions
+from vllm.inputs import MultiModalDataDict
 from vllm.model_executor.layers.linear import ColumnParallelLinear
 from vllm.model_executor.models.persimmon import PersimmonForCausalLM
 from vllm.multimodal import MULTIMODAL_REGISTRY
 from vllm.multimodal.inputs import (
-    MultiModalDataDict,
     MultiModalFieldConfig,
     MultiModalKwargsItems,
 )

vllm/model_executor/models/gemma3_mm.py

@@ -12,12 +12,12 @@ from transformers.models.gemma3.processing_gemma3 import Gemma3ProcessorKwargs
 
 from vllm.config import VllmConfig
 from vllm.config.multimodal import BaseDummyOptions
+from vllm.inputs import MultiModalDataDict
 from vllm.logger import init_logger
 from vllm.model_executor.layers.layernorm import GemmaRMSNorm
 from vllm.model_executor.models.module_mapping import MultiModelKeys
 from vllm.multimodal import MULTIMODAL_REGISTRY
 from vllm.multimodal.inputs import (
-    MultiModalDataDict,
     MultiModalFieldConfig,
     MultiModalKwargsItems,
 )

vllm/model_executor/models/gemma3n_mm.py

@@ -19,7 +19,7 @@ from transformers.models.siglip import SiglipImageProcessorFast
 
 from vllm.config import ModelConfig, SpeechToTextConfig, VllmConfig
 from vllm.config.multimodal import BaseDummyOptions
-from vllm.inputs.data import PromptType, TextPrompt
+from vllm.inputs import MultiModalDataDict, PromptType, TextPrompt
 from vllm.logger import init_logger
 from vllm.model_executor.layers.layernorm import RMSNorm
 from vllm.model_executor.layers.linear import RowParallelLinear
@@ -32,7 +32,6 @@ from vllm.model_executor.models.module_mapping import MultiModelKeys
 from vllm.model_executor.models.whisper import ISO639_1_SUPPORTED_LANGS
 from vllm.multimodal import MULTIMODAL_REGISTRY
 from vllm.multimodal.inputs import (
-    MultiModalDataDict,
     MultiModalFieldConfig,
     MultiModalKwargsItems,
 )

vllm/model_executor/models/glm4_1v.py

@@ -50,6 +50,7 @@ from vllm.config import VllmConfig
 from vllm.config.multimodal import BaseDummyOptions, VideoDummyOptions
 from vllm.distributed import get_tensor_model_parallel_world_size, parallel_state
 from vllm.distributed import utils as dist_utils
+from vllm.inputs import MultiModalDataDict
 from vllm.logger import init_logger
 from vllm.model_executor.layers.attention import (
     MMEncoderAttention,
@@ -74,7 +75,6 @@ from vllm.model_executor.model_loader.weight_utils import default_weight_loader
 from vllm.model_executor.models.module_mapping import MultiModelKeys
 from vllm.multimodal import MULTIMODAL_REGISTRY
 from vllm.multimodal.inputs import (
-    MultiModalDataDict,
     MultiModalFeatureSpec,
     MultiModalFieldConfig,
     MultiModalKwargsItems,

vllm/model_executor/models/glm4v.py

@@ -18,6 +18,7 @@ from transformers import BatchFeature
 from vllm.config import VllmConfig
 from vllm.config.multimodal import BaseDummyOptions
 from vllm.distributed import get_tensor_model_parallel_world_size
+from vllm.inputs import MultiModalDataDict
 from vllm.model_executor.layers.activation import SiluAndMul, get_act_fn
 from vllm.model_executor.layers.attention import MMEncoderAttention
 from vllm.model_executor.layers.conv import Conv2dLayer
@@ -32,7 +33,6 @@ from vllm.model_executor.layers.quantization import QuantizationConfig
 from vllm.model_executor.models.module_mapping import MultiModelKeys
 from vllm.multimodal import MULTIMODAL_REGISTRY
 from vllm.multimodal.inputs import (
-    MultiModalDataDict,
     MultiModalFeatureSpec,
     MultiModalFieldConfig,
     MultiModalKwargsItems,

vllm/model_executor/models/glmasr.py

@@ -14,7 +14,7 @@ from transformers.models.whisper import WhisperFeatureExtractor
 from vllm.config import ModelConfig, SpeechToTextConfig, VllmConfig
 from vllm.config.multimodal import BaseDummyOptions
 from vllm.distributed.parallel_state import get_tensor_model_parallel_world_size
-from vllm.inputs.data import PromptType, TokensPrompt
+from vllm.inputs import ModalityData, MultiModalDataDict, PromptType, TokensPrompt
 from vllm.model_executor.layers.activation import get_act_fn
 from vllm.model_executor.layers.attention import MMEncoderAttention
 from vllm.model_executor.layers.linear import (
@@ -27,13 +27,11 @@ from vllm.model_executor.layers.rotary_embedding.common import ApplyRotaryEmb
 from vllm.model_executor.models.module_mapping import MultiModelKeys
 from vllm.multimodal import MULTIMODAL_REGISTRY
 from vllm.multimodal.inputs import (
-    MultiModalDataDict,
     MultiModalFieldConfig,
     MultiModalKwargsItems,
 )
 from vllm.multimodal.parse import (
     DictEmbeddingItems,
-    ModalityData,
     ModalityDataItems,
     MultiModalDataItems,
     MultiModalDataParser,

vllm/model_executor/models/granite_speech.py

@@ -36,13 +36,12 @@ from transformers import BatchFeature, PretrainedConfig
 
 from vllm.config import CacheConfig, ModelConfig, SpeechToTextConfig, VllmConfig
 from vllm.config.multimodal import BaseDummyOptions
-from vllm.inputs.data import PromptType, TokensPrompt
+from vllm.inputs import MultiModalDataDict, PromptType, TokensPrompt
 from vllm.model_executor.layers.linear import ColumnParallelLinear, RowParallelLinear
 from vllm.model_executor.layers.quantization import QuantizationConfig
 from vllm.model_executor.models.module_mapping import MultiModelKeys
 from vllm.multimodal import MULTIMODAL_REGISTRY
 from vllm.multimodal.inputs import (
-    MultiModalDataDict,
     MultiModalFieldConfig,
     MultiModalKwargsItems,
 )

vllm/model_executor/models/hunyuan_vision.py

@@ -37,6 +37,7 @@ from vllm.config import VllmConfig
 from vllm.config.multimodal import BaseDummyOptions
 from vllm.distributed import parallel_state
 from vllm.distributed import utils as dist_utils
+from vllm.inputs import ModalityData, MultiModalDataDict
 from vllm.logger import init_logger
 from vllm.model_executor.layers.activation import get_act_fn
 from vllm.model_executor.layers.attention import MMEncoderAttention
@@ -52,8 +53,6 @@ from vllm.model_executor.models.module_mapping import MultiModelKeys
 from vllm.multimodal import MULTIMODAL_REGISTRY
 from vllm.multimodal.inputs import (
     ImageItem,
-    ModalityData,
-    MultiModalDataDict,
     MultiModalFeatureSpec,
     MultiModalFieldConfig,
     MultiModalKwargsItems,

vllm/model_executor/models/hyperclovax_vision.py

@@ -18,10 +18,10 @@ from transformers import BatchFeature, CLIPVisionConfig, SiglipVisionConfig
 
 from vllm.config import VllmConfig
 from vllm.config.multimodal import BaseDummyOptions
+from vllm.inputs import MultiModalDataDict
 from vllm.model_executor.layers.quantization import QuantizationConfig
 from vllm.multimodal import MULTIMODAL_REGISTRY
 from vllm.multimodal.inputs import (
-    MultiModalDataDict,
     MultiModalFieldConfig,
     MultiModalKwargsItems,
 )

vllm/model_executor/models/hyperclovax_vision_v2.py

@@ -21,9 +21,9 @@ from transformers import BatchFeature
 from vllm.config import VllmConfig
 from vllm.config.multimodal import BaseDummyOptions
 from vllm.forward_context import set_forward_context
+from vllm.inputs import MultiModalDataDict
 from vllm.multimodal import MULTIMODAL_REGISTRY
 from vllm.multimodal.inputs import (
-    MultiModalDataDict,
     MultiModalFieldConfig,
     MultiModalKwargsItems,
 )

vllm/model_executor/models/idefics3.py

@@ -30,6 +30,7 @@ from transformers import (
 
 from vllm.config import VllmConfig
 from vllm.config.multimodal import BaseDummyOptions
+from vllm.inputs import MultiModalDataDict
 from vllm.model_executor.layers.linear import ReplicatedLinear
 from vllm.model_executor.layers.logits_processor import LogitsProcessor
 from vllm.model_executor.layers.quantization import QuantizationConfig
@@ -37,7 +38,6 @@ from vllm.model_executor.layers.vocab_parallel_embedding import ParallelLMHead
 from vllm.model_executor.models.module_mapping import MultiModelKeys
 from vllm.multimodal import MULTIMODAL_REGISTRY
 from vllm.multimodal.inputs import (
-    MultiModalDataDict,
     MultiModalFieldConfig,
     MultiModalKwargsItems,
 )

vllm/model_executor/models/interfaces.py

@@ -30,8 +30,7 @@ from transformers.models.whisper.tokenization_whisper import LANGUAGES
 from typing_extensions import Self, TypeIs
 
 from vllm.config import ModelConfig, SpeechToTextConfig
-from vllm.inputs import TokensPrompt
-from vllm.inputs.data import PromptType
+from vllm.inputs import PromptType, TokensPrompt
 from vllm.logger import init_logger
 from vllm.model_executor.layers.mamba.mamba_utils import MambaStateCopyFunc
 from vllm.model_executor.layers.quantization import QuantizationConfig

vllm/model_executor/models/interns1.py

@@ -23,12 +23,12 @@ from transformers.models.internvl.video_processing_internvl import (
 
 from vllm.config import VllmConfig
 from vllm.config.multimodal import BaseDummyOptions
+from vllm.inputs import MultiModalDataDict
 from vllm.model_executor.layers.quantization import QuantizationConfig
 from vllm.model_executor.models.interns1_vit import InternS1VisionModel
 from vllm.model_executor.models.module_mapping import MultiModelKeys
 from vllm.multimodal import MULTIMODAL_REGISTRY
 from vllm.multimodal.inputs import (
-    MultiModalDataDict,
     MultiModalFieldConfig,
     MultiModalKwargsItems,
 )

vllm/model_executor/models/internvl.py

@@ -18,6 +18,7 @@ from transformers import BatchFeature, PretrainedConfig
 
 from vllm.config import VllmConfig
 from vllm.config.multimodal import BaseDummyOptions
+from vllm.inputs import MultiModalDataDict
 from vllm.model_executor.layers.quantization import QuantizationConfig
 from vllm.model_executor.layers.quantization.awq import AWQConfig
 from vllm.model_executor.models.intern_vit import (
@@ -28,7 +29,6 @@ from vllm.model_executor.models.module_mapping import MultiModelKeys
 from vllm.multimodal import MULTIMODAL_REGISTRY
 from vllm.multimodal.inputs import (
     BatchedTensorInputs,
-    MultiModalDataDict,
     MultiModalFieldConfig,
     MultiModalKwargsItems,
 )

vllm/model_executor/models/isaac.py

@@ -16,6 +16,7 @@ from vllm.config import ModelConfig, VllmConfig
 from vllm.config.multimodal import BaseDummyOptions
 from vllm.distributed import parallel_state
 from vllm.distributed import utils as dist_utils
+from vllm.inputs import MultiModalDataDict
 from vllm.model_executor.layers.attention import MMEncoderAttention
 from vllm.model_executor.layers.linear import (
     ColumnParallelLinear,
@@ -44,7 +45,6 @@ from vllm.model_executor.models.utils import (
 )
 from vllm.multimodal import MULTIMODAL_REGISTRY
 from vllm.multimodal.inputs import (
-    MultiModalDataDict,
     MultiModalFeatureSpec,
     MultiModalFieldConfig,
     MultiModalKwargsItems,

vllm/model_executor/models/kanana_v.py

@@ -19,10 +19,10 @@ from transformers.models.qwen2_vl.configuration_qwen2_vl import Qwen2VLVisionCon
 
 from vllm.config import VllmConfig
 from vllm.config.multimodal import BaseDummyOptions
+from vllm.inputs import MultiModalDataDict
 from vllm.logger import init_logger
 from vllm.multimodal import MULTIMODAL_REGISTRY
 from vllm.multimodal.inputs import (
-    MultiModalDataDict,
     MultiModalFieldConfig,
     MultiModalKwargsItems,
 )