xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-05-27 07:07:52 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Core] Remove legacy input mapper/processor from V0 (#15686)

Browse Source 
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
This commit is contained in:
Cyrus Leung 2025-04-28 15:38:48 +08:00 committed by GitHub
parent 7fcc4223dc
commit aec9674dbe
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
21 changed files with 142 additions and 1337 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
vllm/core/scheduler.py
View File

@ -1596,7 +1596,6 @@ class Scheduler:
multi_modal_placeholders=( multi_modal_placeholders=(
seq_group.multi_modal_placeholders seq_group.multi_modal_placeholders
if scheduler_outputs.num_prefill_groups > 0 else None), if scheduler_outputs.num_prefill_groups > 0 else None),
mm_processor_kwargs=seq_group.mm_processor_kwargs,
prompt_adapter_request=seq_group.prompt_adapter_request, prompt_adapter_request=seq_group.prompt_adapter_request,
) )
else: else:

3
vllm/engine/async_llm_engine.py
View File

@ -493,12 +493,11 @@ class _AsyncLLMEngine(LLMEngine):
tokenizer = await self.get_tokenizer_async(lora_request) tokenizer = await self.get_tokenizer_async(lora_request)
self._validate_token_prompt(prompt, tokenizer=tokenizer) self._validate_token_prompt(prompt, tokenizer=tokenizer)
preprocessed_inputs = await self.input_preprocessor.preprocess_async( processed_inputs = await self.input_preprocessor.preprocess_async(
prompt, prompt,
lora_request=lora_request, lora_request=lora_request,
prompt_adapter_request=prompt_adapter_request, prompt_adapter_request=prompt_adapter_request,
) )
processed_inputs = self.input_processor(preprocessed_inputs)
if isinstance(params, SamplingParams) and \ if isinstance(params, SamplingParams) and \
params.guided_decoding is not None: params.guided_decoding is not None:

13
vllm/engine/llm_engine.py
View File

@ -29,8 +29,7 @@ from vllm.engine.output_processor.util import create_output_by_sequence_group
from vllm.entrypoints.openai.logits_processors import ( from vllm.entrypoints.openai.logits_processors import (
get_logits_processors as get_openai_logits_processors) get_logits_processors as get_openai_logits_processors)
from vllm.executor.executor_base import ExecutorBase from vllm.executor.executor_base import ExecutorBase
from vllm.inputs import (INPUT_REGISTRY, InputRegistry, ProcessorInputs, from vllm.inputs import ProcessorInputs, PromptType, SingletonInputs
PromptType, SingletonInputs)
from vllm.inputs.parse import is_token_prompt, split_enc_dec_inputs from vllm.inputs.parse import is_token_prompt, split_enc_dec_inputs
from vllm.inputs.preprocess import InputPreprocessor from vllm.inputs.preprocess import InputPreprocessor
from vllm.logger import init_logger from vllm.logger import init_logger
@ -213,7 +212,6 @@ class LLMEngine:
log_stats: bool, log_stats: bool,
usage_context: UsageContext = UsageContext.ENGINE_CONTEXT, usage_context: UsageContext = UsageContext.ENGINE_CONTEXT,
stat_loggers: Optional[Dict[str, StatLoggerBase]] = None, stat_loggers: Optional[Dict[str, StatLoggerBase]] = None,
input_registry: InputRegistry = INPUT_REGISTRY,
mm_registry: MultiModalRegistry = MULTIMODAL_REGISTRY, mm_registry: MultiModalRegistry = MULTIMODAL_REGISTRY,
use_cached_outputs: bool = False, use_cached_outputs: bool = False,
) -> None: ) -> None:
@ -274,11 +272,7 @@ class LLMEngine:
self.tokenizer, self.tokenizer,
mm_registry) mm_registry)
self.input_registry = input_registry self.model_executor = executor_class(vllm_config=vllm_config)
self.input_processor = input_registry.create_input_processor(
self.model_config)
self.model_executor = executor_class(vllm_config=vllm_config, )
if self.model_config.runner_type != "pooling": if self.model_config.runner_type != "pooling":
self._initialize_kv_caches() self._initialize_kv_caches()
@ -762,12 +756,11 @@ class LLMEngine:
prompt, prompt,
tokenizer=self.get_tokenizer(lora_request=lora_request)) tokenizer=self.get_tokenizer(lora_request=lora_request))
preprocessed_inputs = self.input_preprocessor.preprocess( processed_inputs = self.input_preprocessor.preprocess(
prompt, prompt,
lora_request=lora_request, lora_request=lora_request,
prompt_adapter_request=prompt_adapter_request, prompt_adapter_request=prompt_adapter_request,
) )
processed_inputs = self.input_processor(preprocessed_inputs)
self._add_processed_request( self._add_processed_request(
request_id=request_id, request_id=request_id,

8
vllm/inputs/__init__.py
View File

@ -2,10 +2,9 @@
from .data import (DecoderOnlyInputs, EncoderDecoderInputs, from .data import (DecoderOnlyInputs, EncoderDecoderInputs,
ExplicitEncoderDecoderPrompt, ProcessorInputs, PromptType, ExplicitEncoderDecoderPrompt, ProcessorInputs, PromptType,
SingletonInputs, SingletonInputsAdapter, SingletonPrompt, SingletonInputs, SingletonPrompt, TextPrompt, TokenInputs,
TextPrompt, TokenInputs, TokensPrompt, TokensPrompt, build_explicit_enc_dec_prompt,
build_explicit_enc_dec_prompt, to_enc_dec_tuple_list, to_enc_dec_tuple_list, token_inputs, zip_enc_dec_prompts)
token_inputs, zip_enc_dec_prompts)
from .registry import (DummyData, InputContext, InputProcessingContext, from .registry import (DummyData, InputContext, InputProcessingContext,
InputRegistry) InputRegistry)
@ -27,7 +26,6 @@ __all__ = [
"EncoderDecoderInputs", "EncoderDecoderInputs",
"ProcessorInputs", "ProcessorInputs",
"SingletonInputs", "SingletonInputs",
"SingletonInputsAdapter",
"build_explicit_enc_dec_prompt", "build_explicit_enc_dec_prompt",
"to_enc_dec_tuple_list", "to_enc_dec_tuple_list",
"zip_enc_dec_prompts", "zip_enc_dec_prompts",

161
vllm/inputs/data.py
View File

@ -1,17 +1,11 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
from collections.abc import Iterable from collections.abc import Iterable
from dataclasses import dataclass
from functools import cached_property
from typing import TYPE_CHECKING, Any, Generic, Literal, Optional, Union, cast from typing import TYPE_CHECKING, Any, Generic, Literal, Optional, Union, cast
import torch from typing_extensions import NotRequired, TypedDict, TypeVar
from typing_extensions import NotRequired, TypedDict, TypeVar, assert_never
if TYPE_CHECKING: if TYPE_CHECKING:
from vllm.multimodal import (MultiModalDataDict, MultiModalKwargs, from vllm.multimodal.inputs import MultiModalDataDict, MultiModalInputs
MultiModalPlaceholderDict)
from vllm.multimodal.inputs import MultiModalInputs
class TextPrompt(TypedDict): class TextPrompt(TypedDict):
@ -147,46 +141,11 @@ class TokenInputs(TypedDict):
The original prompt text corresponding to the token IDs, if available. The original prompt text corresponding to the token IDs, if available.
""" """
multi_modal_data: NotRequired["MultiModalDataDict"]
"""
Optional multi-modal data to pass to the model,
if the model supports it.
"""
multi_modal_inputs: NotRequired["MultiModalKwargs"]
"""
Optional multi-modal inputs to pass to the model,
if the model supports it.
"""
multi_modal_placeholders: NotRequired["MultiModalPlaceholderDict"]
"""
Placeholder ranges for the multi-modal data.
"""
multi_modal_hashes: NotRequired[list[str]]
"""
The hashes of the multi-modal data.
"""
mm_processor_kwargs: NotRequired[dict[str, Any]]
"""
Optional multi-modal processor kwargs to be forwarded to the
multimodal input mapper & processor. Note that if multiple modalities
have registered mappers etc for the model being considered, we attempt
to pass the mm_processor_kwargs to each of them.
"""
def token_inputs( def token_inputs(
prompt_token_ids: list[int], prompt_token_ids: list[int],
token_type_ids: Optional[list[int]] = None, token_type_ids: Optional[list[int]] = None,
prompt: Optional[str] = None, prompt: Optional[str] = None,
multi_modal_data: Optional["MultiModalDataDict"] = None,
multi_modal_inputs: Optional["MultiModalKwargs"] = None,
multi_modal_hashes: Optional[list[str]] = None,
multi_modal_placeholders: Optional["MultiModalPlaceholderDict"] = None,
mm_processor_kwargs: Optional[dict[str, Any]] = None,
) -> TokenInputs: ) -> TokenInputs:
"""Construct :class:`TokenInputs` from optional values.""" """Construct :class:`TokenInputs` from optional values."""
inputs = TokenInputs(type="token", prompt_token_ids=prompt_token_ids) inputs = TokenInputs(type="token", prompt_token_ids=prompt_token_ids)
@ -195,16 +154,6 @@ def token_inputs(
inputs["prompt"] = prompt inputs["prompt"] = prompt
if token_type_ids is not None: if token_type_ids is not None:
inputs["token_type_ids"] = token_type_ids inputs["token_type_ids"] = token_type_ids
if multi_modal_data is not None:
inputs["multi_modal_data"] = multi_modal_data
if multi_modal_inputs is not None:
inputs["multi_modal_inputs"] = multi_modal_inputs
if multi_modal_hashes is not None:
inputs["multi_modal_hashes"] = multi_modal_hashes
if multi_modal_placeholders is not None:
inputs["multi_modal_placeholders"] = multi_modal_placeholders
if mm_processor_kwargs is not None:
inputs["mm_processor_kwargs"] = mm_processor_kwargs
return inputs return inputs
@ -237,112 +186,6 @@ A processed :class:`SingletonPrompt` which can be passed to
:class:`vllm.sequence.Sequence`. :class:`vllm.sequence.Sequence`.
""" """
@dataclass
class SingletonInputsAdapter:
"""
Unified interface to access the components of :class:`SingletonInputs`.
"""
inputs: SingletonInputs
@cached_property
def prompt(self) -> Optional[str]:
inputs = self.inputs
if inputs["type"] == "token" or inputs["type"] == "multimodal":
return inputs.get("prompt")
assert_never(inputs) # type: ignore[arg-type]
@cached_property
def prompt_token_ids(self) -> list[int]:
inputs = self.inputs
if inputs["type"] == "token" or inputs["type"] == "multimodal":
return inputs.get("prompt_token_ids", [])
assert_never(inputs) # type: ignore[arg-type]
@cached_property
def token_type_ids(self) -> list[int]:
inputs = self.inputs
if inputs["type"] == "token" or inputs["type"] == "multimodal":
return inputs.get("token_type_ids", [])
assert_never(inputs) # type: ignore[arg-type]
@cached_property
def prompt_embeds(self) -> Optional[torch.Tensor]:
inputs = self.inputs
if inputs["type"] == "token" or inputs["type"] == "multimodal":
return None
assert_never(inputs) # type: ignore[arg-type]
@cached_property
def multi_modal_data(self) -> "MultiModalDataDict":
inputs = self.inputs
if inputs["type"] == "token":
return inputs.get("multi_modal_data", {})
if inputs["type"] == "multimodal":
return inputs.get("mm_kwargs", {})
assert_never(inputs) # type: ignore[arg-type]
@cached_property
def multi_modal_inputs(self) -> Union[dict, "MultiModalKwargs"]:
inputs = self.inputs
if inputs["type"] == "token":
return inputs.get("multi_modal_inputs", {})
if inputs["type"] == "multimodal":
return inputs.get("mm_kwargs", {})
assert_never(inputs) # type: ignore[arg-type]
@cached_property
def multi_modal_hashes(self) -> list[str]:
inputs = self.inputs
if inputs["type"] == "token":
return inputs.get("multi_modal_hashes", [])
if inputs["type"] == "multimodal":
# only the case when we use MultiModalInputs
return inputs.get("mm_hashes", []) # type: ignore[return-value]
assert_never(inputs) # type: ignore[arg-type]
@cached_property
def multi_modal_placeholders(self) -> "MultiModalPlaceholderDict":
inputs = self.inputs
if inputs["type"] == "token":
return inputs.get("multi_modal_placeholders", {})
if inputs["type"] == "multimodal":
return inputs.get("mm_placeholders", {})
assert_never(inputs) # type: ignore[arg-type]
@cached_property
def mm_processor_kwargs(self) -> dict[str, Any]:
inputs = self.inputs
if inputs["type"] == "token":
return inputs.get("mm_processor_kwargs", {})
if inputs["type"] == "multimodal":
return {}
assert_never(inputs) # type: ignore[arg-type]
ProcessorInputs = Union[DecoderOnlyInputs, EncoderDecoderInputs] ProcessorInputs = Union[DecoderOnlyInputs, EncoderDecoderInputs]
""" """
The inputs to :data:`vllm.inputs.InputProcessor`. The inputs to :data:`vllm.inputs.InputProcessor`.

60
vllm/inputs/preprocess.py
View File

@ -223,28 +223,6 @@ class InputPreprocessor:
lora_request=lora_request, lora_request=lora_request,
add_special_tokens=add_special_tokens) add_special_tokens=add_special_tokens)
def _can_process_multimodal(self) -> bool:
model_config = self.model_config
if not model_config.is_multimodal_model:
raise ValueError("Your model does not support multi-modal inputs")
# Interim measure so we can handle models that have yet to be
# updated to use the new multi-modal processor
can_process_multimodal = self.mm_registry.has_processor(model_config)
if not can_process_multimodal:
from vllm.model_executor.models.registry import _VLLM_MODELS
if not any(arch in _VLLM_MODELS
for arch in model_config.architectures):
logger.warning_once(
"Your model uses the legacy input pipeline, which will be "
"removed in an upcoming release. "
"Please upgrade to the new multi-modal processing pipeline "
"(https://docs.vllm.ai/en/latest/design/mm_processing.html)"
)
return can_process_multimodal
def _process_multimodal( def _process_multimodal(
self, self,
prompt: Union[str, list[int]], prompt: Union[str, list[int]],
@ -258,8 +236,7 @@ class InputPreprocessor:
returning the corresponding token IDs and metadata. returning the corresponding token IDs and metadata.
""" """
# At the moment on model (PrithviGeoSpatialMAE) requires to be # At the moment on model (PrithviGeoSpatialMAE) requires to be
# initialized without a tokenizer while using also multi-modal # initialized without a tokenizer while using also multi-modal input
# input.
if not self.tokenizer: if not self.tokenizer:
tokenizer = object() # Dummy tokenizer = object() # Dummy
else: else:
@ -285,8 +262,7 @@ class InputPreprocessor:
) -> MultiModalInputs: ) -> MultiModalInputs:
"""Async version of :meth:`_process_multimodal`.""" """Async version of :meth:`_process_multimodal`."""
# At the moment on model (PrithviGeoSpatialMAE) requires to be # At the moment on model (PrithviGeoSpatialMAE) requires to be
# initialized without a tokenizer while using also multi-modal # initialized without a tokenizer while using also multi-modal input
# input.
if not self.tokenizer: if not self.tokenizer:
tokenizer = object() # Dummy tokenizer = object() # Dummy
else: else:
@ -343,7 +319,7 @@ class InputPreprocessor:
multi_modal_data = tokens_content.get("multi_modal_data") multi_modal_data = tokens_content.get("multi_modal_data")
mm_processor_kwargs = tokens_content.get("mm_processor_kwargs") mm_processor_kwargs = tokens_content.get("mm_processor_kwargs")
if multi_modal_data is not None and self._can_process_multimodal(): if multi_modal_data is not None:
return self._process_multimodal( return self._process_multimodal(
prompt_token_ids, prompt_token_ids,
multi_modal_data, multi_modal_data,
@ -355,8 +331,6 @@ class InputPreprocessor:
return token_inputs( return token_inputs(
prompt_token_ids=prompt_token_ids, prompt_token_ids=prompt_token_ids,
token_type_ids=token_type_ids, token_type_ids=token_type_ids,
multi_modal_data=multi_modal_data,
mm_processor_kwargs=mm_processor_kwargs,
) )
if parsed["type"] == "text": if parsed["type"] == "text":
@ -366,7 +340,7 @@ class InputPreprocessor:
multi_modal_data = text_content.get("multi_modal_data") multi_modal_data = text_content.get("multi_modal_data")
mm_processor_kwargs = text_content.get("mm_processor_kwargs") mm_processor_kwargs = text_content.get("mm_processor_kwargs")
if multi_modal_data is not None and self._can_process_multimodal(): if multi_modal_data is not None:
return self._process_multimodal( return self._process_multimodal(
prompt_text, prompt_text,
multi_modal_data, multi_modal_data,
@ -383,8 +357,6 @@ class InputPreprocessor:
return token_inputs( return token_inputs(
prompt=prompt_text, prompt=prompt_text,
prompt_token_ids=prompt_token_ids, prompt_token_ids=prompt_token_ids,
multi_modal_data=multi_modal_data,
mm_processor_kwargs=mm_processor_kwargs,
) )
assert_never(parsed) assert_never(parsed)
@ -417,7 +389,7 @@ class InputPreprocessor:
multi_modal_data = tokens_content.get("multi_modal_data") multi_modal_data = tokens_content.get("multi_modal_data")
mm_processor_kwargs = tokens_content.get("mm_processor_kwargs") mm_processor_kwargs = tokens_content.get("mm_processor_kwargs")
if multi_modal_data is not None and self._can_process_multimodal(): if multi_modal_data is not None:
return await self._process_multimodal_async( return await self._process_multimodal_async(
prompt_token_ids, prompt_token_ids,
multi_modal_data, multi_modal_data,
@ -426,11 +398,7 @@ class InputPreprocessor:
return_mm_hashes=return_mm_hashes, return_mm_hashes=return_mm_hashes,
) )
return token_inputs( return token_inputs(prompt_token_ids=prompt_token_ids)
prompt_token_ids=prompt_token_ids,
multi_modal_data=multi_modal_data,
mm_processor_kwargs=mm_processor_kwargs,
)
if parsed["type"] == "text": if parsed["type"] == "text":
text_content = parsed["content"] text_content = parsed["content"]
@ -439,7 +407,7 @@ class InputPreprocessor:
multi_modal_data = text_content.get("multi_modal_data") multi_modal_data = text_content.get("multi_modal_data")
mm_processor_kwargs = text_content.get("mm_processor_kwargs") mm_processor_kwargs = text_content.get("mm_processor_kwargs")
if multi_modal_data is not None and self._can_process_multimodal(): if multi_modal_data is not None:
return await self._process_multimodal_async( return await self._process_multimodal_async(
prompt_text, prompt_text,
multi_modal_data, multi_modal_data,
@ -456,8 +424,6 @@ class InputPreprocessor:
return token_inputs( return token_inputs(
prompt=prompt_text, prompt=prompt_text,
prompt_token_ids=prompt_token_ids, prompt_token_ids=prompt_token_ids,
multi_modal_data=multi_modal_data,
mm_processor_kwargs=mm_processor_kwargs,
) )
assert_never(parsed) assert_never(parsed)
@ -594,15 +560,13 @@ class InputPreprocessor:
decoder_inputs = self._prompt_to_llm_inputs(decoder_input) decoder_inputs = self._prompt_to_llm_inputs(decoder_input)
# For multimodal model, override decoder prompt from processor # For multimodal model, override decoder prompt from processor
# with explicit decoder prompt. # with explicit decoder prompt.
if self.model_config.is_multimodal_model and ( if self.model_config.is_multimodal_model:
self._can_process_multimodal()):
encoder_inputs, decoder_inputs = ( encoder_inputs, decoder_inputs = (
self._separate_enc_dec_inputs_from_mm_processor_outputs( self._separate_enc_dec_inputs_from_mm_processor_outputs(
encoder_inputs, decoder_inputs)) encoder_inputs, decoder_inputs))
else: else:
inputs = self._prompt_to_llm_inputs(prompt) inputs = self._prompt_to_llm_inputs(prompt)
if self.model_config.is_multimodal_model and ( if self.model_config.is_multimodal_model:
self._can_process_multimodal()):
# Encoder-Decoder Multimodal model # Encoder-Decoder Multimodal model
encoder_inputs, decoder_inputs = ( encoder_inputs, decoder_inputs = (
self._separate_enc_dec_inputs_from_mm_processor_outputs( self._separate_enc_dec_inputs_from_mm_processor_outputs(
@ -637,15 +601,13 @@ class InputPreprocessor:
# For multimodal model, override decoder prompt from processor # For multimodal model, override decoder prompt from processor
# with explicit decoder prompt. # with explicit decoder prompt.
if self.model_config.is_multimodal_model and ( if self.model_config.is_multimodal_model:
self._can_process_multimodal()):
encoder_inputs, decoder_inputs = ( encoder_inputs, decoder_inputs = (
self._separate_enc_dec_inputs_from_mm_processor_outputs( self._separate_enc_dec_inputs_from_mm_processor_outputs(
encoder_inputs, decoder_inputs)) encoder_inputs, decoder_inputs))
else: else:
inputs = await self._prompt_to_llm_inputs_async(prompt) inputs = await self._prompt_to_llm_inputs_async(prompt)
if self.model_config.is_multimodal_model and ( if self.model_config.is_multimodal_model:
self._can_process_multimodal()):
# Encoder-Decoder Multimodal model # Encoder-Decoder Multimodal model
encoder_inputs, decoder_inputs = ( encoder_inputs, decoder_inputs = (
self._separate_enc_dec_inputs_from_mm_processor_outputs( self._separate_enc_dec_inputs_from_mm_processor_outputs(

321
vllm/inputs/registry.py
View File

@ -1,24 +1,14 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
import functools
from collections import UserDict
from collections.abc import Mapping from collections.abc import Mapping
from dataclasses import dataclass from dataclasses import dataclass
from typing import (TYPE_CHECKING, Any, Callable, NamedTuple, Optional, from typing import TYPE_CHECKING, Any, NamedTuple, Optional, Union
Protocol, Union)
from torch import nn
from transformers import BatchFeature, PretrainedConfig, ProcessorMixin from transformers import BatchFeature, PretrainedConfig, ProcessorMixin
from typing_extensions import TypeVar, assert_never from typing_extensions import TypeVar
from vllm.logger import init_logger
from vllm.transformers_utils.processor import cached_processor_from_config from vllm.transformers_utils.processor import cached_processor_from_config
from vllm.transformers_utils.tokenizer import AnyTokenizer from vllm.transformers_utils.tokenizer import AnyTokenizer
from vllm.utils import (ClassRegistry, get_allowed_kwarg_only_overrides, from vllm.utils import resolve_mm_processor_kwargs
resolve_mm_processor_kwargs)
from .data import ProcessorInputs, SingletonInputs
from .parse import split_enc_dec_inputs
if TYPE_CHECKING: if TYPE_CHECKING:
from vllm.config import ModelConfig from vllm.config import ModelConfig
@ -26,8 +16,6 @@ if TYPE_CHECKING:
MultiModalRegistry) MultiModalRegistry)
from vllm.sequence import SequenceData from vllm.sequence import SequenceData
logger = init_logger(__name__)
_T = TypeVar("_T") _T = TypeVar("_T")
_C = TypeVar("_C", bound=PretrainedConfig, default=PretrainedConfig) _C = TypeVar("_C", bound=PretrainedConfig, default=PretrainedConfig)
_P = TypeVar("_P", bound=ProcessorMixin, default=ProcessorMixin) _P = TypeVar("_P", bound=ProcessorMixin, default=ProcessorMixin)
@ -172,142 +160,23 @@ class InputProcessingContext(InputContext):
raise RuntimeError(msg) from exc raise RuntimeError(msg) from exc
N = TypeVar("N", bound=type[nn.Module])
class DummyData(NamedTuple): class DummyData(NamedTuple):
"""Dummy data used for profiling.""" """
Dummy data used for profiling.
Note: This is only used in V0.
"""
seq_data: "SequenceData" seq_data: "SequenceData"
multi_modal_data: Optional["MultiModalDataDict"] = None multi_modal_data: Optional["MultiModalDataDict"] = None
multi_modal_placeholders: Optional["MultiModalPlaceholderDict"] = None multi_modal_placeholders: Optional["MultiModalPlaceholderDict"] = None
class DummyDataFactory(Protocol):
def __call__(
self,
ctx: InputContext,
seq_len: int,
mm_counts: Mapping[str, int],
**mm_processor_kwargs: Any,
) -> DummyData:
"""
Create dummy data to be inputted into the model.
Note:
:data:`InputProcessor` is not applied to the dummy data.
The :code:`mm_processor_kwargs` are overrides provided at
initialization time to values in the config whose values
may affect the number of tokens per instance.
"""
...
class _MultiModalCounts(UserDict[str, int]):
"""
Wraps `mm_counts` for a more informative error message
when attempting to access a plugin that does not exist.
"""
def __getitem__(self, key: str) -> int:
try:
return super().__getitem__(key)
except KeyError as exc:
msg = (f"There is no multi-modal plugin with the key: {key}. "
f"Available keys: {set(self.keys())}")
raise KeyError(msg) from exc
InputProcessor = Callable[[InputContext, ProcessorInputs], ProcessorInputs]
"""Preprocess the inputs to the model."""
class InputRegistry: class InputRegistry:
""" """
A registry to dispatch data processing Note: This is only used in V0.
according to the target model.
""" """
def __init__(self) -> None:
self._dummy_factories_by_model_type = \
ClassRegistry[nn.Module, DummyDataFactory]()
self._dummy_encoder_factories_by_model_type = \
ClassRegistry[nn.Module, DummyDataFactory]()
self._input_processors_by_model_type = \
ClassRegistry[nn.Module, InputProcessor]()
def _default_dummy_data_factory(
self,
ctx: InputContext,
seq_len: int,
mm_counts: Mapping[str, int],
) -> DummyData:
"""
The default dummy data factory represents the longest possible text
that can be inputted to the model.
Note:
:data:`InputProcessor` is not applied to the dummy data.
"""
# Avoid circular import
from vllm.sequence import SequenceData
return DummyData(SequenceData.from_prompt_token_counts((0, seq_len)))
def register_dummy_data(self, factory: DummyDataFactory):
"""
Register a dummy data factory to a model class.
During memory profiling, the provided function is invoked to create
dummy data to be inputted into the model. The resulting memory usage
should be an upper bound of what the model would use at inference time.
"""
def wrapper(model_cls: N) -> N:
if self._dummy_factories_by_model_type.contains(model_cls,
strict=True):
logger.warning(
"Model class %s already has dummy data "
"registered to %s. It is overwritten by the new one.",
model_cls, self)
self._dummy_factories_by_model_type[model_cls] = factory
return model_cls
return wrapper
def _get_dummy_data_factory(self, model_cls: type[nn.Module]):
return self._dummy_factories_by_model_type \
.get(model_cls, self._default_dummy_data_factory)
def register_dummy_encoder_data(self, factory: DummyDataFactory):
"""
Register a dummy encoder data factory to a model class
This is similar to :meth:`~register_dummy_data`, but for encoder input.
"""
def wrapper(model_cls: N) -> N:
if self._dummy_encoder_factories_by_model_type.contains(
model_cls, strict=True):
logger.warning(
"Model class %s already has dummy encoder data "
"registered to %s. It is overwritten by the new one.",
model_cls, self)
self._dummy_encoder_factories_by_model_type[model_cls] = factory
return model_cls
return wrapper
def _get_dummy_encoder_data_factory(self, model_cls: type[nn.Module]):
return self._dummy_encoder_factories_by_model_type \
.get(model_cls, self._default_dummy_data_factory)
def dummy_data_for_profiling( def dummy_data_for_profiling(
self, self,
model_config: "ModelConfig", model_config: "ModelConfig",
@ -319,169 +188,25 @@ class InputRegistry:
Create dummy data for profiling the memory usage of a model. Create dummy data for profiling the memory usage of a model.
The model is identified by ``model_config``. The model is identified by ``model_config``.
Note:
This should be called after
:meth:`~MultiModalRegistry.init_mm_limits_per_prompt`.
""" """
# Avoid circular import # Avoid circular import
from vllm.model_executor.model_loader import get_model_architecture
from vllm.multimodal import MultiModalKwargs
from vllm.multimodal.profiling import MultiModalProfiler
from vllm.sequence import SequenceData from vllm.sequence import SequenceData
if mm_registry.has_processor(model_config): if not model_config.is_multimodal_model:
processor = mm_registry.create_processor(model_config, seq_data = SequenceData.from_prompt_token_counts((0, seq_len))
disable_cache=True) return DummyData(seq_data=seq_data)
profiler = MultiModalProfiler(processor)
dummy_data_v1 = (profiler.get_encoder_dummy_data(seq_len) # Encoder dummy data does not contain multi-modal data
if is_encoder_data else if is_encoder_data:
profiler.get_decoder_dummy_data(seq_len)) enc_data = mm_registry.get_encoder_dummy_data(
_seq_data = SequenceData.from_seqs( model_config, seq_len)
dummy_data_v1.prompt_token_ids) # type: ignore[attr-defined] seq_data = SequenceData.from_seqs(enc_data.prompt_token_ids)
return DummyData(seq_data=seq_data)
dummy_data = DummyData( dec_data = mm_registry.get_decoder_dummy_data(model_config, seq_len)
seq_data=_seq_data,
multi_modal_data=getattr(dummy_data_v1, "multi_modal_data",
None),
multi_modal_placeholders=getattr(dummy_data_v1,
"multi_modal_placeholders",
None),
)
else:
model_cls, _ = get_model_architecture(model_config)
if is_encoder_data:
dummy_factory = self._get_dummy_encoder_data_factory(model_cls)
else:
dummy_factory = self._get_dummy_data_factory(model_cls)
mm_counts = mm_registry.get_mm_limits_per_prompt(model_config)
mm_processor_kwargs = get_allowed_kwarg_only_overrides(
dummy_factory,
overrides=model_config.mm_processor_kwargs,
requires_kw_only=False,
allow_var_kwargs=True,
)
dummy_data = dummy_factory(InputContext(model_config), seq_len, return DummyData(
_MultiModalCounts(mm_counts), seq_data=SequenceData.from_seqs(dec_data.prompt_token_ids),
**mm_processor_kwargs) multi_modal_data=dec_data.multi_modal_data,
multi_modal_placeholders=dec_data.multi_modal_placeholders,
# Having more tokens is over-conservative but otherwise fine
num_tokens = dummy_data.seq_data.prompt_token_ids
if len(num_tokens) < seq_len:
if is_encoder_data:
logger.warning_once(
f"Expected at least {seq_len} dummy encoder tokens for "
f"profiling, but found {len(num_tokens)} tokens instead.")
else:
raise AssertionError(
f"Expected at least {seq_len} dummy tokens for profiling, "
f"but found {len(num_tokens)} tokens instead.")
if (dummy_data.multi_modal_data is not None and
not isinstance(dummy_data.multi_modal_data, MultiModalKwargs)):
for k, v in dummy_data.multi_modal_data.items():
num_items = len(v) if isinstance(v, list) else 1
num_expected = mm_counts[k]
assert num_items >= num_expected, (
f"Expected at least {num_expected} dummy '{k}' instances "
f"for profiling, but found {num_items} instances instead.")
return dummy_data
def _default_input_processor(
self,
ctx: InputContext,
inputs: ProcessorInputs,
**kwargs: object,
) -> ProcessorInputs:
"""The default input processor is a no-op."""
return inputs
def register_input_processor(self, processor: InputProcessor):
"""
Register an input processor to a model class.
The provided function is invoked on each input to the model. This
happens before
:meth:`~vllm.multimodal.registry.MultiModalRegistry.map_input`.
"""
def wrapper(model_cls: N) -> N:
if self._input_processors_by_model_type.contains(model_cls,
strict=True):
logger.warning(
"Model class %s already has input processor "
"registered to %s. It is overwritten by the new one.",
model_cls, self)
self._input_processors_by_model_type[model_cls] = processor
return model_cls
return wrapper
def _get_model_input_processor(self, model_cls: type[nn.Module]):
return self._input_processors_by_model_type \
.get(model_cls, self._default_input_processor)
def _ensure_mm_kwargs(
self,
inputs: SingletonInputs,
mm_processor_kwargs: dict[str, Any],
):
if inputs["type"] == "token":
# In case the input processor for that model fails to set it
if "mm_processor_kwargs" not in inputs:
inputs["mm_processor_kwargs"] = mm_processor_kwargs
elif inputs["type"] == "multimodal":
# Be more strict in V2
assert "mm_kwargs" in inputs
else:
assert_never(inputs["type"]) # type: ignore[arg-type]
def process_input(self, model_config: "ModelConfig",
inputs: ProcessorInputs) -> ProcessorInputs:
"""
Apply an input processor to an instance of model inputs.
The model is identified by ``model_config``.
"""
# Avoid circular import
from vllm.model_executor.model_loader import get_model_architecture
model_cls, _ = get_model_architecture(model_config)
processor = self._get_model_input_processor(model_cls)
# Handle multimodal processor kwargs with priority:
# Inference kwargs -> Init kwargs -> {}
# If it's empty, it'll fall back to the default kwarg values
mm_processor_kwargs = resolve_mm_processor_kwargs(
model_config.mm_processor_kwargs,
inputs.get("mm_processor_kwargs", {}), # type: ignore
processor,
requires_kw_only=False,
allow_var_kwargs=True,
) )
processed_inputs = processor(
InputContext(model_config),
inputs,
**mm_processor_kwargs,
)
encoder_inputs, decoder_inputs = split_enc_dec_inputs(processed_inputs)
if encoder_inputs is not None:
self._ensure_mm_kwargs(encoder_inputs, mm_processor_kwargs)
if decoder_inputs is not None:
self._ensure_mm_kwargs(decoder_inputs, mm_processor_kwargs)
return processed_inputs
def create_input_processor(self, model_config: "ModelConfig"):
"""
Create an input processor (see :meth:`_process_input`) for a
specific model.
"""
return functools.partial(self.process_input, model_config)

4
vllm/multimodal/__init__.py
View File

@ -1,6 +1,5 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
from .base import MultiModalPlaceholderMap
from .base import MultiModalPlaceholderMap, MultiModalPlugin
from .hasher import MultiModalHashDict, MultiModalHasher from .hasher import MultiModalHashDict, MultiModalHasher
from .inputs import (BatchedTensorInputs, ModalityData, MultiModalDataBuiltins, from .inputs import (BatchedTensorInputs, ModalityData, MultiModalDataBuiltins,
MultiModalDataDict, MultiModalKwargs, MultiModalDataDict, MultiModalKwargs,
@ -26,7 +25,6 @@ __all__ = [
"MultiModalKwargs", "MultiModalKwargs",
"MultiModalPlaceholderDict", "MultiModalPlaceholderDict",
"MultiModalPlaceholderMap", "MultiModalPlaceholderMap",
"MultiModalPlugin",
"NestedTensors", "NestedTensors",
"MULTIMODAL_REGISTRY", "MULTIMODAL_REGISTRY",
"MultiModalRegistry", "MultiModalRegistry",

23
vllm/multimodal/audio.py
View File

@ -7,11 +7,9 @@ from typing import Literal, Optional
import numpy as np import numpy as np
import numpy.typing as npt import numpy.typing as npt
from vllm.inputs.registry import InputContext
from vllm.utils import PlaceholderModule from vllm.utils import PlaceholderModule
from .base import MediaIO, MultiModalPlugin from .base import MediaIO
from .inputs import AudioItem, ModalityData, MultiModalKwargs
try: try:
import librosa import librosa
@ -24,25 +22,6 @@ except ImportError:
soundfile = PlaceholderModule("soundfile") # type: ignore[assignment] soundfile = PlaceholderModule("soundfile") # type: ignore[assignment]
class AudioPlugin(MultiModalPlugin):
"""Plugin for audio data."""
def get_data_key(self) -> str:
return "audio"
def _default_input_mapper(
self,
ctx: InputContext,
data: ModalityData[AudioItem],
**mm_processor_kwargs,
) -> MultiModalKwargs:
raise NotImplementedError("There is no default audio input mapper")
def _default_max_multimodal_tokens(self, ctx: InputContext) -> int:
raise NotImplementedError(
"There is no default maximum multimodal tokens")
def resample_audio_librosa( def resample_audio_librosa(
audio: npt.NDArray[np.floating], audio: npt.NDArray[np.floating],
*, *,

284
vllm/multimodal/base.py
View File

@ -1,247 +1,23 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
from abc import ABC, abstractmethod from abc import ABC, abstractmethod
from collections import defaultdict
from collections.abc import Sequence from collections.abc import Sequence
from pathlib import Path from pathlib import Path
from typing import (TYPE_CHECKING, Any, Callable, Generic, NamedTuple, from typing import TYPE_CHECKING, Generic, NamedTuple, TypeVar
Optional, TypeVar, Union)
from torch import nn
from vllm.inputs import InputContext
from vllm.logger import init_logger
from vllm.utils import (ClassRegistry, get_allowed_kwarg_only_overrides,
resolve_mm_processor_kwargs)
if TYPE_CHECKING: if TYPE_CHECKING:
from vllm.config import ModelConfig
from vllm.sequence import SequenceGroupMetadata from vllm.sequence import SequenceGroupMetadata
from .inputs import (ModalityData, MultiModalDataDict, MultiModalKwargs, from .inputs import MultiModalKwargs, PlaceholderRange
PlaceholderRange)
logger = init_logger(__name__)
MultiModalInputMapper = Callable[[InputContext, ModalityData[object]],
MultiModalKwargs]
"""
Return a dictionary to be passed as keyword arguments to
:meth:`~torch.nn.Module.forward`. This is similar in concept to tokenizers
and processors in HuggingFace Transformers.
If the data is not supported, throw :exc:`TypeError`.
"""
MultiModalTokensCalc = Union[int, Callable[[InputContext], int]]
"""
Calculate the maximum number of multimodal tokens input to the language
model. This does not include tokens that correspond to the input text.
"""
_T = TypeVar("_T") _T = TypeVar("_T")
N = TypeVar("N", bound=type[nn.Module])
class MultiModalPlugin(ABC):
"""
Base class that defines data processing logic for a specific modality.
In particular, we adopt a registry pattern to dispatch data processing
according to the model being used (considering that different models may
process the same data differently). This registry is in turn used by
:class:`~MultiModalRegistry` which acts at a higher level
(i.e., the modality of the data).
"""
def __init__(self) -> None:
self._input_mappers = ClassRegistry[nn.Module, MultiModalInputMapper]()
self._max_mm_tokens = ClassRegistry[nn.Module, MultiModalTokensCalc]()
@abstractmethod
def get_data_key(self) -> str:
"""
Get the data key corresponding to the modality.
"""
raise NotImplementedError
@abstractmethod
def _default_input_mapper(
self,
ctx: InputContext,
data: ModalityData[Any],
**mm_processor_kwargs,
) -> MultiModalKwargs:
"""
Return a dictionary to be passed as keyword arguments to
:meth:`~torch.nn.Module.forward`. This is similar in concept to
tokenizers and processors in HuggingFace Transformers.
If the data is not supported, throw :exc:`TypeError`.
"""
raise NotImplementedError
def register_input_mapper(
self,
mapper: Optional[MultiModalInputMapper] = None,
):
"""
Register an input mapper to a model class.
When the model receives input data that matches the modality served by
this plugin (see :meth:`get_data_key`), the provided function is
invoked to transform the data into a dictionary of model inputs.
If `None` is provided, then the default input mapper is used instead.
"""
def wrapper(model_cls: N) -> N:
if self._input_mappers.contains(model_cls, strict=True):
logger.warning(
"Model class %s already has an input mapper "
"registered to %s. It is overwritten by the new one.",
model_cls,
self,
)
self._input_mappers[model_cls] = (mapper
or self._default_input_mapper)
return model_cls
return wrapper
def map_input(
self,
model_config: "ModelConfig",
data: ModalityData[Any],
mm_processor_kwargs: Optional[dict[str, Any]],
) -> MultiModalKwargs:
"""
Transform the data into a dictionary of model inputs using the
input mapper registered for that model.
The model is identified by ``model_config``.
Raises:
TypeError: If the data type is not supported.
"""
# Avoid circular import
from vllm.model_executor.model_loader import get_model_architecture
model_cls, _ = get_model_architecture(model_config)
mapper = self._input_mappers.get(model_cls)
if mapper is None:
raise KeyError(f"No input mapper in {self} is registered for "
f"model class {model_cls.__name__}.")
if mm_processor_kwargs is None:
mm_processor_kwargs = {}
# In the case of the default mapper, we have to get resource
# processor through its HuggingFace autoclass; since this goes
# through **kwargs, we can't inspect it the same way, so we allow
# drop mm_processor_kwargs based on signature inspection
# if we're using the default mapper.
#
# This should be safe in general due to the sanitation, since the
# transformers resource should filter unused kwargs anyway.
uses_default_mapper = mapper == self._default_input_mapper
mm_processor_kwargs = resolve_mm_processor_kwargs(
model_config.mm_processor_kwargs,
mm_processor_kwargs,
callable=mapper,
allow_var_kwargs=uses_default_mapper,
)
return mapper(InputContext(model_config), data, **mm_processor_kwargs)
@abstractmethod
def _default_max_multimodal_tokens(self, ctx: InputContext) -> int:
"""
Calculate the maximum number of tokens, corresponding to a single
instance of multimodal data, that are passed to the language model.
"""
raise NotImplementedError
def _validate_max_multimodal_tokens(self, max_mm_tokens: int):
if max_mm_tokens < 1:
raise ValueError("You should set the number of tokens to a "
f"positive integer. Found: {max_mm_tokens}")
def register_max_multimodal_tokens(
self,
max_mm_tokens: Optional[MultiModalTokensCalc] = None,
):
"""
Register the maximum number of tokens, corresponding to a single
instance of multimodal data, that are passed to the language model
for a model class.
If `None` is provided, then the default calculation is used instead.
"""
def wrapper(model_cls: N) -> N:
if self._max_mm_tokens.contains(model_cls, strict=True):
logger.warning(
"Model class %s already calculates maximum number of "
"tokens in %s. It is overwritten by the new one.",
model_cls,
self,
)
if isinstance(max_mm_tokens, int):
self._validate_max_multimodal_tokens(max_mm_tokens)
self._max_mm_tokens[model_cls] = (
max_mm_tokens or self._default_max_multimodal_tokens)
return model_cls
return wrapper
def get_max_multimodal_tokens(self, model_config: "ModelConfig") -> int:
"""
Get the maximum number of multi-modal tokens
for profiling the memory usage of a model.
If this registry is not applicable to the model, `0` is returned.
The model is identified by ``model_config``.
"""
# Avoid circular import
from vllm.model_executor.model_loader import get_model_architecture
from vllm.model_executor.models import supports_multimodal
model_cls, _ = get_model_architecture(model_config)
if not supports_multimodal(model_cls):
return 0
max_mm_tokens = self._max_mm_tokens.get(model_cls)
if max_mm_tokens is None:
return 0
if callable(max_mm_tokens):
mm_processor_kwargs = get_allowed_kwarg_only_overrides(
max_mm_tokens,
overrides=model_config.mm_processor_kwargs,
requires_kw_only=False,
allow_var_kwargs=True,
)
max_mm_tokens = max_mm_tokens(InputContext(model_config),
**mm_processor_kwargs)
self._validate_max_multimodal_tokens(max_mm_tokens)
return max_mm_tokens
class MultiModalPlaceholderMap: class MultiModalPlaceholderMap:
""" """
Relates multi-modal embeddings to their corresponding placeholders. Relates multi-modal embeddings to their corresponding placeholders.
Note: This is only used in V0.
""" """
class IndexMap(NamedTuple): class IndexMap(NamedTuple):
@ -279,8 +55,7 @@ class MultiModalPlaceholderMap:
@classmethod @classmethod
def from_seq_group( def from_seq_group(
cls, seq_group: "SequenceGroupMetadata", positions: range cls, seq_group: "SequenceGroupMetadata", positions: range
) -> tuple[Optional[MultiModalDataDict], dict[str, ) -> tuple[MultiModalKwargs, dict[str, "MultiModalPlaceholderMap"]]:
"MultiModalPlaceholderMap"]]:
""" """
Returns the multi-modal items that intersect with the portion of a Returns the multi-modal items that intersect with the portion of a
prompt (``seq_group``) represented by ``positions``, as well as a prompt (``seq_group``) represented by ``positions``, as well as a
@ -323,48 +98,24 @@ class MultiModalPlaceholderMap:
seq_mm_placeholders = seq_group.multi_modal_placeholders seq_mm_placeholders = seq_group.multi_modal_placeholders
if not seq_mm_data or not seq_mm_placeholders: if not seq_mm_data or not seq_mm_placeholders:
return seq_mm_data, {} return MultiModalKwargs({}), {}
# For merged processor, we directly use mm_kwargs as mm_data placeholder_maps = dict[str, MultiModalPlaceholderMap]()
if isinstance(seq_mm_data, MultiModalKwargs):
placeholder_maps = dict[str, MultiModalPlaceholderMap]()
for modality, placeholders in seq_mm_placeholders.items():
placeholder_map = MultiModalPlaceholderMap()
if positions:
placeholder_map.append_items_from_seq_group(
positions,
# Dummy, since we don't care about intersecting items
[None] * len(placeholders),
placeholders,
)
placeholder_maps[modality] = placeholder_map
return seq_mm_data, placeholder_maps
mm_data = {**seq_mm_data}
placeholder_maps = defaultdict[str, MultiModalPlaceholderMap](
MultiModalPlaceholderMap)
for modality, placeholders in seq_mm_placeholders.items(): for modality, placeholders in seq_mm_placeholders.items():
mm_items = mm_data.pop(modality) placeholder_map = MultiModalPlaceholderMap()
if not isinstance(mm_items, list):
mm_items = [mm_items]
if positions: if positions:
intersecting_items = placeholder_maps[modality] \ placeholder_map.append_items_from_seq_group(
.append_items_from_seq_group( positions,
positions, # Dummy, since we don't care about intersecting items
mm_items, [None] * len(placeholders),
placeholders, placeholders,
) )
if intersecting_items: placeholder_maps[modality] = placeholder_map
mm_data[modality] = intersecting_items
return mm_data, placeholder_maps return seq_mm_data, placeholder_maps
def append_items_from_seq_group( def append_items_from_seq_group(
self, self,
@ -445,8 +196,7 @@ class MultiModalPlaceholderMap:
f"The number of source ({len(src_indices)}) and destination " f"The number of source ({len(src_indices)}) and destination "
f"indices ({len(dest_indices)}) must be the same.") f"indices ({len(dest_indices)}) must be the same.")
return MultiModalPlaceholderMap.IndexMap(src=src_indices, return self.IndexMap(src=src_indices, dest=dest_indices)
dest=dest_indices)
class MediaIO(ABC, Generic[_T]): class MediaIO(ABC, Generic[_T]):

80
vllm/multimodal/image.py
View File

@ -3,89 +3,11 @@
import base64 import base64
from io import BytesIO from io import BytesIO
from pathlib import Path from pathlib import Path
from typing import TYPE_CHECKING, Any, Optional
import torch import torch
from PIL import Image from PIL import Image
from vllm.inputs.registry import InputContext from .base import MediaIO
from vllm.logger import init_logger
from vllm.transformers_utils.processor import cached_get_image_processor
from vllm.utils import is_list_of
from .base import MediaIO, MultiModalPlugin
from .inputs import ImageItem, ModalityData, MultiModalKwargs
if TYPE_CHECKING:
from vllm.config import ModelConfig
logger = init_logger(__name__)
class ImagePlugin(MultiModalPlugin):
"""Plugin for image data."""
def get_data_key(self) -> str:
return "image"
def _get_hf_image_processor(
self,
model_config: "ModelConfig",
mm_processor_kwargs: Optional[dict[str, Any]] = None,
):
if mm_processor_kwargs is None:
mm_processor_kwargs = {}
return cached_get_image_processor(
model_config.model,
trust_remote_code=model_config.trust_remote_code,
**mm_processor_kwargs)
def _default_input_mapper(
self,
ctx: InputContext,
data: ModalityData[ImageItem],
**mm_processor_kwargs,
) -> MultiModalKwargs:
model_config = ctx.model_config
# PIL image
if isinstance(data, Image.Image) or is_list_of(data, Image.Image):
image_processor = self._get_hf_image_processor(
model_config,
mm_processor_kwargs,
)
if image_processor is None:
raise RuntimeError("No HuggingFace processor is available "
"to process the image object")
try:
# NOTE: It may make sense to forward the mm_processor_kwargs
# here too. For now, to keep it simple, we only allow it be
# used for the initialization call though, just in case the
# signatures of the preprocessor initializer don't match
# preprocess()
batch_data = image_processor \
.preprocess(data, return_tensors="pt") \
.data
except Exception:
logger.error(
"Failed to process image (%s) with the default mapper. "
"This is most likely an edge-case with this model's image "
"processor in transformers (type: %s), and not vLLM.",
data,
type(image_processor).__name__)
raise
return MultiModalKwargs(batch_data)
# Image embedding
elif isinstance(data, torch.Tensor) or is_list_of(data, torch.Tensor):
return MultiModalKwargs({"image_embeds": data})
raise TypeError(f"Invalid image type: {type(data)}")
def _default_max_multimodal_tokens(self, ctx: InputContext) -> int:
return 3000
def rescale_image_size(image: Image.Image, def rescale_image_size(image: Image.Image,

271
vllm/multimodal/registry.py
View File

@ -1,13 +1,10 @@
# SPDX-License-Identifier: Apache-2.0 # SPDX-License-Identifier: Apache-2.0
from collections.abc import Mapping
import functools
import json
from collections import UserDict
from collections.abc import Mapping, Sequence
from dataclasses import dataclass from dataclasses import dataclass
from typing import TYPE_CHECKING, Any, Generic, Optional, Protocol, TypeVar from typing import TYPE_CHECKING, Generic, Optional, Protocol, TypeVar
import torch.nn as nn import torch.nn as nn
from typing_extensions import deprecated
from vllm.envs import VLLM_MM_INPUT_CACHE_GIB from vllm.envs import VLLM_MM_INPUT_CACHE_GIB
from vllm.inputs import InputProcessingContext from vllm.inputs import InputProcessingContext
@ -16,15 +13,10 @@ from vllm.transformers_utils.tokenizer import (AnyTokenizer,
cached_tokenizer_from_config) cached_tokenizer_from_config)
from vllm.utils import ClassRegistry from vllm.utils import ClassRegistry
from .audio import AudioPlugin
from .base import MultiModalInputMapper, MultiModalPlugin, MultiModalTokensCalc
from .image import ImagePlugin
from .inputs import MultiModalDataDict, MultiModalKwargs, NestedTensors
from .processing import (BaseMultiModalProcessor, BaseProcessingInfo, from .processing import (BaseMultiModalProcessor, BaseProcessingInfo,
ProcessingCache) ProcessingCache)
from .profiling import (BaseDummyInputsBuilder, DummyDecoderData, from .profiling import (BaseDummyInputsBuilder, DummyDecoderData,
DummyEncoderData, MultiModalProfiler) DummyEncoderData, MultiModalProfiler)
from .video import VideoPlugin
if TYPE_CHECKING: if TYPE_CHECKING:
from vllm.config import ModelConfig from vllm.config import ModelConfig
@ -85,169 +77,23 @@ class _ProcessorFactories(Generic[_I]):
return self.processor(info, dummy_inputs_builder, cache=cache) return self.processor(info, dummy_inputs_builder, cache=cache)
class _MultiModalLimits(UserDict["ModelConfig", dict[str, int]]):
"""
Wraps `_limits_by_model` for a more informative error message
when attempting to access a model that does not exist.
"""
def __getitem__(self, key: "ModelConfig") -> dict[str, int]:
try:
return super().__getitem__(key)
except KeyError as exc:
msg = (f"Cannot find `mm_limits` for model={key.model}. Did you "
"forget to call `init_mm_limits_per_prompt`?")
raise KeyError(msg) from exc
class MultiModalRegistry: class MultiModalRegistry:
""" """
A registry that dispatches data processing according to the model. A registry that dispatches data processing according to the model.
""" """
DEFAULT_PLUGINS = (ImagePlugin(), AudioPlugin(), VideoPlugin()) def __init__(self) -> None:
def __init__(
self,
*,
plugins: Sequence[MultiModalPlugin] = DEFAULT_PLUGINS) -> None:
self._plugins = {p.get_data_key(): p for p in plugins}
self._processor_factories = ClassRegistry[nn.Module, self._processor_factories = ClassRegistry[nn.Module,
_ProcessorFactories]() _ProcessorFactories]()
# This is used for non-multimodal models
self._disabled_limits_per_plugin = {k: 0 for k in self._plugins}
self._limits_by_model = _MultiModalLimits()
self._processing_cache = ProcessingCache(VLLM_MM_INPUT_CACHE_GIB) self._processing_cache = ProcessingCache(VLLM_MM_INPUT_CACHE_GIB)
def register_plugin(self, plugin: MultiModalPlugin) -> None: @deprecated("Legacy input processor/mapper pipeline has been removed. "
""" "Please update your model runner to use "
Register a multi-modal plugin so it can be recognized by vLLM. "`seq_group_metadata.multi_modal_data` directly without "
""" "further processing.")
data_type_key = plugin.get_data_key()
if data_type_key in self._plugins:
logger.warning(
"A plugin is already registered for data type %s, "
"and will be overwritten by the new plugin %s.", data_type_key,
plugin)
self._plugins[data_type_key] = plugin
def _get_plugin(self, data_type_key: str):
plugin = self._plugins.get(data_type_key)
if plugin is not None:
return plugin
msg = f"Unknown multi-modal data type: {data_type_key}"
raise NotImplementedError(msg)
def register_input_mapper(
self,
data_type_key: str,
mapper: Optional[MultiModalInputMapper] = None,
):
"""
Register an input mapper for a specific modality to a model class.
See :meth:`MultiModalPlugin.register_input_mapper` for more details.
"""
return self._get_plugin(data_type_key).register_input_mapper(mapper)
def register_image_input_mapper(
self,
mapper: Optional[MultiModalInputMapper] = None,
):
"""
Register an input mapper for image data to a model class.
See :meth:`MultiModalPlugin.register_input_mapper` for more details.
"""
return self.register_input_mapper("image", mapper)
def map_input(
self,
model_config: "ModelConfig",
data: MultiModalDataDict,
mm_processor_kwargs: Optional[dict[str, Any]] = None,
) -> MultiModalKwargs:
"""
Apply an input mapper to the data passed to the model.
The data belonging to each modality is passed to the corresponding
plugin which in turn converts the data into into keyword arguments
via the input mapper registered for that model.
See :meth:`MultiModalPlugin.map_input` for more details.
Note:
This should be called after :meth:`init_mm_limits_per_prompt`.
"""
merged_dict = dict[str, NestedTensors]()
for data_key, data_value in data.items():
plugin = self._get_plugin(data_key)
num_items = len(data_value) if isinstance(data_value, list) else 1
max_items = self._limits_by_model[model_config][data_key]
if num_items > max_items:
raise ValueError(
f"You set '{json.dumps({data_key: max_items})}' (or "
"defaulted to 1) in `--limit-mm-per-prompt`, but found "
f"{num_items} items in the same prompt.")
input_dict = plugin.map_input(model_config, data_value,
mm_processor_kwargs)
for input_key, input_tensor in input_dict.items():
if input_key in merged_dict:
raise ValueError(f"The input mappers (keys={set(data)}) "
f"resulted in a conflicting keyword "
f"argument to `forward()`: {input_key}")
merged_dict[input_key] = input_tensor
return MultiModalKwargs(merged_dict)
def create_input_mapper(self, model_config: "ModelConfig"): def create_input_mapper(self, model_config: "ModelConfig"):
""" return lambda data, mm_processor_kwargs: data
Create an input mapper (see :meth:`map_input`) for a specific model.
"""
# NOTE - we currently make the assumption that if a model has multiple
# supported modalities, they take the same kwargs. For the default,
# this could be an issue in the future if it falls back to two HF
# resources and we can't inspect the signature easily since it's
# getting initialized through the autoclass.
#
# If this is a problem in the future, we should revisit it, but since
# it potentially introduces a lot of complexity for a currently
# uncommon case, we do not for simplicity of both use & implementation
return functools.partial(self.map_input, model_config)
def register_max_multimodal_tokens(
self,
data_type_key: str,
max_mm_tokens: Optional[MultiModalTokensCalc] = None,
):
"""
Register the maximum number of tokens, corresponding to a single
instance of multimodal data belonging to a specific modality, that are
passed to the language model for a model class.
"""
return self._get_plugin(data_type_key) \
.register_max_multimodal_tokens(max_mm_tokens)
def register_max_image_tokens(
self,
max_mm_tokens: Optional[MultiModalTokensCalc] = None,
):
"""
Register the maximum number of image tokens, corresponding to a single
image, that are passed to the language model for a model class.
"""
return self.register_max_multimodal_tokens("image", max_mm_tokens)
def get_max_tokens_per_item_by_modality( def get_max_tokens_per_item_by_modality(
self, self,
@ -257,25 +103,22 @@ class MultiModalRegistry:
Get the maximum number of tokens per data item from each modality based Get the maximum number of tokens per data item from each modality based
on underlying model configuration. on underlying model configuration.
""" """
if self.has_processor(model_config): if not model_config.is_multimodal_model:
processor = self.create_processor(model_config, disable_cache=True) return {}
profiler = MultiModalProfiler(processor)
seq_len = model_config.max_model_len processor = self.create_processor(model_config, disable_cache=True)
mm_limits = self.get_mm_limits_per_prompt(model_config) profiler = MultiModalProfiler(processor)
return profiler.get_mm_max_tokens( seq_len = model_config.max_model_len
seq_len, mm_limits = self.get_mm_limits_per_prompt(model_config)
{
modality: 1
for modality, limit in mm_limits.items() if limit > 0
},
)
return { return profiler.get_mm_max_tokens(
key: plugin.get_max_multimodal_tokens(model_config) seq_len,
for key, plugin in self._plugins.items() {
} modality: 1
for modality, limit in mm_limits.items() if limit > 0
},
)
def get_max_tokens_per_item_by_nonzero_modality( def get_max_tokens_per_item_by_nonzero_modality(
self, self,
@ -308,9 +151,6 @@ class MultiModalRegistry:
for profiling the memory usage of a model. for profiling the memory usage of a model.
See :meth:`MultiModalPlugin.get_max_multimodal_tokens` for more details. See :meth:`MultiModalPlugin.get_max_multimodal_tokens` for more details.
Note:
This should be called after :meth:`init_mm_limits_per_prompt`.
""" """
mm_limits = self.get_mm_limits_per_prompt(model_config) mm_limits = self.get_mm_limits_per_prompt(model_config)
@ -326,47 +166,18 @@ class MultiModalRegistry:
for profiling the memory usage of a model. for profiling the memory usage of a model.
See :meth:`MultiModalPlugin.get_max_multimodal_tokens` for more details. See :meth:`MultiModalPlugin.get_max_multimodal_tokens` for more details.
Note:
This should be called after :meth:`init_mm_limits_per_prompt`.
""" """
return sum(self.get_max_tokens_by_modality(model_config).values()) return sum(self.get_max_tokens_by_modality(model_config).values())
@deprecated("Legacy input processor/mapper pipeline has been removed. "
"Please update your model runner to use "
"`seq_group_metadata.multi_modal_data` directly without "
"further processing.")
def init_mm_limits_per_prompt( def init_mm_limits_per_prompt(
self, self,
model_config: "ModelConfig", model_config: "ModelConfig",
) -> None: ) -> None:
""" pass
Initialize the maximum number of multi-modal input instances for each
modality that are allowed per prompt for a model class.
"""
if model_config in self._limits_by_model:
logger.warning(
"`mm_limits` has already been set for model=%s, and will "
"be overwritten by the new values.", model_config.model)
multimodal_config = model_config.multimodal_config
if multimodal_config is None:
limits_per_plugin = self._disabled_limits_per_plugin
else:
config_limits_per_plugin = multimodal_config.limit_per_prompt
extra_keys = config_limits_per_plugin.keys() - self._plugins.keys()
if extra_keys:
logger.warning(
"Detected extra keys in `--limit-mm-per-prompt` which "
"are not registered as multi-modal plugins: %s. "
"They will be ignored.", extra_keys)
# NOTE: Currently the default is set to 1 for each plugin
# TODO: Automatically determine the limits based on budget
# once more models support multi-image inputs
limits_per_plugin = {
key: multimodal_config.get_limit_per_prompt(key)
for key in self._plugins
}
self._limits_by_model[model_config] = limits_per_plugin
def get_mm_limits_per_prompt( def get_mm_limits_per_prompt(
self, self,
@ -375,16 +186,13 @@ class MultiModalRegistry:
""" """
Get the maximum number of multi-modal input instances for each modality Get the maximum number of multi-modal input instances for each modality
that are allowed per prompt for a model class. that are allowed per prompt for a model class.
Note:
This should be called after :meth:`init_mm_limits_per_prompt`.
""" """
if self.has_processor(model_config): if not model_config.is_multimodal_model:
processor = self.create_processor(model_config, disable_cache=True) return {}
profiler = MultiModalProfiler(processor)
return profiler.get_mm_limits()
return self._limits_by_model[model_config] processor = self.create_processor(model_config, disable_cache=True)
profiler = MultiModalProfiler(processor)
return profiler.get_mm_limits()
def register_processor( def register_processor(
self, self,
@ -428,14 +236,12 @@ class MultiModalRegistry:
model_cls, _ = get_model_architecture(model_config) model_cls, _ = get_model_architecture(model_config)
return model_cls return model_cls
@deprecated("Legacy input processor/mapper pipeline has been removed. "
"Please update your model runner to use "
"`seq_group_metadata.multi_modal_data` directly without "
"further processing.")
def has_processor(self, model_config: "ModelConfig") -> bool: def has_processor(self, model_config: "ModelConfig") -> bool:
""" return True
Test whether a multi-modal processor is defined for a specific model.
See also:
:ref:`mm-processing`
"""
return self._get_model_cls(model_config) in self._processor_factories
def create_processor( def create_processor(
self, self,
@ -450,6 +256,9 @@ class MultiModalRegistry:
See also: See also:
:ref:`mm-processing` :ref:`mm-processing`
""" """
if not model_config.is_multimodal_model:
raise ValueError(f"{model_config.model} is not a multimodal model")
if tokenizer is None: if tokenizer is None:
tokenizer = cached_tokenizer_from_config(model_config) tokenizer = cached_tokenizer_from_config(model_config)
if disable_cache is None: if disable_cache is None:

71
vllm/multimodal/video.py
View File

@ -4,80 +4,13 @@ import base64
from functools import partial from functools import partial
from io import BytesIO from io import BytesIO
from pathlib import Path from pathlib import Path
from typing import TYPE_CHECKING, Any, Optional
import numpy as np import numpy as np
import numpy.typing as npt import numpy.typing as npt
from PIL import Image from PIL import Image
from vllm.inputs.registry import InputContext from .base import MediaIO
from vllm.logger import init_logger from .image import ImageMediaIO
from vllm.transformers_utils.processor import cached_get_video_processor
from vllm.utils import is_list_of
from .base import MediaIO, ModalityData
from .image import ImageMediaIO, ImagePlugin
from .inputs import MultiModalKwargs, VideoItem
if TYPE_CHECKING:
from vllm.config import ModelConfig
logger = init_logger(__name__)
class VideoPlugin(ImagePlugin):
"""Plugin for video data."""
def get_data_key(self) -> str:
return "video"
def _get_hf_video_processor(
self,
model_config: "ModelConfig",
mm_processor_kwargs: Optional[dict[str, Any]] = None,
):
if mm_processor_kwargs is None:
mm_processor_kwargs = {}
return cached_get_video_processor(
model_config.model,
trust_remote_code=model_config.trust_remote_code,
**mm_processor_kwargs)
def _default_input_mapper(
self,
ctx: InputContext,
data: ModalityData[VideoItem],
**mm_processor_kwargs,
) -> MultiModalKwargs:
model_config = ctx.model_config
if isinstance(data, list) and len(data) == 1:
data = data[0] # type: ignore
if isinstance(data, np.ndarray) or is_list_of(data, np.ndarray):
video_processor = self._get_hf_video_processor(
model_config,
mm_processor_kwargs,
)
if video_processor is None:
raise RuntimeError("No HuggingFace processor is available "
"to process the video object")
try:
# NOTE: Similar to image; it may be a good idea to filter and
# pass mm_processor_kwargs here too, but for now we don't to
# avoid extra complexity if the initializer and preprocess
# signatures of the processor don't align
batch_data = video_processor(data, return_tensors="pt").data
except Exception:
logger.error("Failed to process video (%s)", data)
raise
return MultiModalKwargs(batch_data)
raise TypeError(f"Invalid video type: {type(data)}")
def _default_max_multimodal_tokens(self, ctx: InputContext) -> int:
return 4096
def resize_video(frames: npt.NDArray, size: tuple[int, int]) -> npt.NDArray: def resize_video(frames: npt.NDArray, size: tuple[int, int]) -> npt.NDArray:

47
vllm/sequence.py
View File

@ -14,9 +14,9 @@ from typing import Any, Callable, Optional, Union
import msgspec import msgspec
import torch import torch
from vllm.inputs import SingletonInputs, SingletonInputsAdapter from vllm.inputs import SingletonInputs
from vllm.lora.request import LoRARequest from vllm.lora.request import LoRARequest
from vllm.multimodal import MultiModalDataDict, MultiModalPlaceholderDict from vllm.multimodal import MultiModalKwargs, MultiModalPlaceholderDict
from vllm.pooling_params import PoolingParams from vllm.pooling_params import PoolingParams
from vllm.prompt_adapter.request import PromptAdapterRequest from vllm.prompt_adapter.request import PromptAdapterRequest
from vllm.sampling_params import RequestOutputKind, SamplingParams from vllm.sampling_params import RequestOutputKind, SamplingParams
@ -419,7 +419,7 @@ class Sequence:
prompt_adapter_request: Optional[PromptAdapterRequest] = None, prompt_adapter_request: Optional[PromptAdapterRequest] = None,
) -> None: ) -> None:
self.seq_id = seq_id self.seq_id = seq_id
self.inputs = SingletonInputsAdapter(inputs) self.inputs = inputs
self.block_size = block_size self.block_size = block_size
self.eos_token_id = eos_token_id self.eos_token_id = eos_token_id
self.lora_request = lora_request self.lora_request = lora_request
@ -448,31 +448,29 @@ class Sequence:
@property @property
def prompt(self) -> Optional[str]: def prompt(self) -> Optional[str]:
return self.inputs.prompt return self.inputs.get("prompt")
@property @property
def prompt_token_ids(self) -> list[int]: def prompt_token_ids(self) -> list[int]:
return self.inputs.prompt_token_ids return self.inputs["prompt_token_ids"]
@property
def prompt_embeds(self) -> Optional[torch.Tensor]:
return self.inputs.prompt_embeds
@property @property
def token_type_ids(self) -> list[int]: def token_type_ids(self) -> list[int]:
return self.inputs.token_type_ids return self.inputs.get("token_type_ids", [])
@property @property
def multi_modal_data(self) -> "MultiModalDataDict": def multi_modal_data(self) -> MultiModalKwargs:
return self.inputs.multi_modal_data if self.inputs["type"] == "multimodal":
return self.inputs["mm_kwargs"]
return MultiModalKwargs({})
@property @property
def multi_modal_placeholders(self) -> MultiModalPlaceholderDict: def multi_modal_placeholders(self) -> MultiModalPlaceholderDict:
return self.inputs.multi_modal_placeholders if self.inputs["type"] == "multimodal":
return self.inputs["mm_placeholders"]
@property return {}
def mm_processor_kwargs(self) -> dict[str, Any]:
return self.inputs.mm_processor_kwargs
@property @property
def lora_int_id(self) -> int: def lora_int_id(self) -> int:
@ -723,12 +721,12 @@ class SequenceGroup:
return self.first_seq.token_type_ids return self.first_seq.token_type_ids
@property @property
def multi_modal_data(self) -> MultiModalDataDict: def multi_modal_data(self) -> MultiModalKwargs:
if self.first_seq.multi_modal_data: if self.first_seq.multi_modal_data:
return self.first_seq.multi_modal_data return self.first_seq.multi_modal_data
elif self.encoder_seq is not None: elif self.encoder_seq is not None:
return self.encoder_seq.multi_modal_data return self.encoder_seq.multi_modal_data
return {} return MultiModalKwargs({})
@property @property
def multi_modal_placeholders(self) -> MultiModalPlaceholderDict: def multi_modal_placeholders(self) -> MultiModalPlaceholderDict:
@ -738,14 +736,6 @@ class SequenceGroup:
return self.encoder_seq.multi_modal_placeholders return self.encoder_seq.multi_modal_placeholders
return {} return {}
@property
def mm_processor_kwargs(self) -> dict[str, Any]:
if self.first_seq.multi_modal_data:
return self.first_seq.mm_processor_kwargs
elif self.encoder_seq is not None:
return self.encoder_seq.mm_processor_kwargs
return {}
@property @property
def lora_int_id(self) -> int: def lora_int_id(self) -> int:
return self.lora_request.lora_int_id if self.lora_request else 0 return self.lora_request.lora_int_id if self.lora_request else 0
@ -969,12 +959,9 @@ class SequenceGroupMetadata(
computed_block_nums: Optional[list[int]] = None computed_block_nums: Optional[list[int]] = None
state: Optional[SequenceGroupState] = msgspec.field( state: Optional[SequenceGroupState] = msgspec.field(
default_factory=lambda: SequenceGroupState()) default_factory=lambda: SequenceGroupState())
# "MultiModalDataDict" types. We have to use Any due to msgspec
# doesn't allow to have union of 2 different dicts.
token_type_ids: Optional[list[int]] = None token_type_ids: Optional[list[int]] = None
multi_modal_data: Optional[Any] = None multi_modal_data: Optional[MultiModalKwargs] = None
multi_modal_placeholders: Optional[MultiModalPlaceholderDict] = None multi_modal_placeholders: Optional[MultiModalPlaceholderDict] = None
mm_processor_kwargs: Optional[dict[str, Any]] = None
encoder_seq_data: Optional[SequenceData] = None encoder_seq_data: Optional[SequenceData] = None
cross_block_table: Optional[list[int]] = None cross_block_table: Optional[list[int]] = None
prompt_adapter_request: Optional[PromptAdapterRequest] = None prompt_adapter_request: Optional[PromptAdapterRequest] = None

35
vllm/transformers_utils/processor.py
View File

@ -208,38 +208,3 @@ def cached_image_processor_from_config(
trust_remote_code=model_config.trust_remote_code, trust_remote_code=model_config.trust_remote_code,
**_merge_mm_kwargs(model_config, **kwargs), **_merge_mm_kwargs(model_config, **kwargs),
) )
def get_video_processor(
processor_name: str,
*args: Any,
trust_remote_code: bool = False,
**kwargs: Any,
):
"""Load a video processor for the given model name via HuggingFace."""
# don't put this import at the top level
# it will call torch.cuda.device_count()
from transformers.image_processing_utils import BaseImageProcessor
processor = get_processor(
processor_name,
*args,
trust_remote_code=trust_remote_code,
**kwargs,
)
return cast(BaseImageProcessor, processor.video_processor)
cached_get_video_processor = lru_cache(get_video_processor)
def cached_video_processor_from_config(
model_config: "ModelConfig",
**kwargs: Any,
):
return cached_get_video_processor(
model_config.model,
trust_remote_code=model_config.trust_remote_code,
**_merge_mm_kwargs(model_config, **kwargs),
)

25
vllm/worker/cpu_model_runner.py
View File

@ -22,8 +22,8 @@ from vllm.model_executor.layers.rotary_embedding import MRotaryEmbedding
from vllm.model_executor.layers.sampler import SamplerOutput, get_sampler from vllm.model_executor.layers.sampler import SamplerOutput, get_sampler
from vllm.model_executor.model_loader import get_model from vllm.model_executor.model_loader import get_model
from vllm.model_executor.models import supports_lora, supports_multimodal from vllm.model_executor.models import supports_lora, supports_multimodal
from vllm.multimodal import (MULTIMODAL_REGISTRY, BatchedTensorInputs, from vllm.multimodal import (BatchedTensorInputs, MultiModalKwargs,
MultiModalKwargs, MultiModalPlaceholderMap) MultiModalPlaceholderMap)
from vllm.sequence import (IntermediateTensors, SequenceData, from vllm.sequence import (IntermediateTensors, SequenceData,
SequenceGroupMetadata) SequenceGroupMetadata)
from vllm.worker.model_runner_base import ( from vllm.worker.model_runner_base import (
@ -154,7 +154,6 @@ class ModelInputForCPUBuilder(ModelRunnerInputBuilderBase[ModelInputForCPU]):
self.sliding_window = self.runner.sliding_window self.sliding_window = self.runner.sliding_window
self.block_size = self.runner.block_size self.block_size = self.runner.block_size
self.device = self.runner.device self.device = self.runner.device
self.multi_modal_input_mapper = self.runner.multi_modal_input_mapper
self.enable_lora = self.runner.lora_config is not None self.enable_lora = self.runner.lora_config is not None
if self.runner.attn_backend is not None: if self.runner.attn_backend is not None:
# spec decode (e.g. Medusa) does not have atten backend # spec decode (e.g. Medusa) does not have atten backend
@ -359,22 +358,14 @@ class ModelInputForCPUBuilder(ModelRunnerInputBuilderBase[ModelInputForCPU]):
computed_len = seq_data.get_num_computed_tokens() computed_len = seq_data.get_num_computed_tokens()
seq_len = self.input_data.seq_lens[-1] seq_len = self.input_data.seq_lens[-1]
# NOTE: mm_data only includes the subset of multi-modal items that # NOTE: mm_kwargs only includes the subset of multi-modal items that
# intersect with the current prefill positions. # intersect with the current prefill positions.
mm_data, placeholder_maps = MultiModalPlaceholderMap.from_seq_group( mm_kwargs, placeholder_maps = MultiModalPlaceholderMap.from_seq_group(
seq_group_metadata, range(computed_len, seq_len)) seq_group_metadata, range(computed_len, seq_len))
if not mm_data: if not mm_kwargs:
return return
if self.runner.mm_registry.has_processor(self.runner.model_config):
mm_kwargs = mm_data
else:
mm_kwargs = self.multi_modal_input_mapper(
mm_data,
seq_group_metadata.mm_processor_kwargs,
)
# special processing for mrope position deltas. # special processing for mrope position deltas.
if self.runner.model_config.uses_mrope: if self.runner.model_config.uses_mrope:
assert not self.chunked_prefill, \ assert not self.chunked_prefill, \
@ -480,12 +471,6 @@ class CPUModelRunnerBase(ModelRunnerBase[TModelInputForCPU]):
use_mla=self.model_config.use_mla, use_mla=self.model_config.use_mla,
) if needs_attn_backend else None ) if needs_attn_backend else None
# Multi-modal data support
self.mm_registry = MULTIMODAL_REGISTRY
self.multi_modal_input_mapper = self.mm_registry \
.create_input_mapper(self.model_config)
self.mm_registry.init_mm_limits_per_prompt(self.model_config)
# Lazy initialization. # Lazy initialization.
self.model: nn.Module # Set after init_Model self.model: nn.Module # Set after init_Model
# Set after load_model. # Set after load_model.

2
vllm/worker/enc_dec_model_runner.py
View File

@ -100,6 +100,8 @@ class EncoderDecoderModelRunner(GPUModelRunnerBase[EncoderDecoderModelInput]):
vllm_config=vllm_config, vllm_config=vllm_config,
kv_cache_dtype=kv_cache_dtype, kv_cache_dtype=kv_cache_dtype,
is_driver_worker=is_driver_worker, is_driver_worker=is_driver_worker,
input_registry=input_registry,
mm_registry=mm_registry,
) )
# Crash for unsupported encoder/scenarios # Crash for unsupported encoder/scenarios

12
vllm/worker/hpu_model_runner.py
View File

@ -45,8 +45,7 @@ from vllm.model_executor.layers.vocab_parallel_embedding import (
VocabParallelEmbedding) VocabParallelEmbedding)
from vllm.model_executor.model_loader import get_model from vllm.model_executor.model_loader import get_model
from vllm.model_executor.sampling_metadata import SequenceGroupToSample from vllm.model_executor.sampling_metadata import SequenceGroupToSample
from vllm.multimodal import (MULTIMODAL_REGISTRY, BatchedTensorInputs, from vllm.multimodal import BatchedTensorInputs, MultiModalKwargs
MultiModalKwargs)
from vllm.sampling_params import SamplingParams from vllm.sampling_params import SamplingParams
from vllm.sequence import (CompletionSequenceGroupOutput, IntermediateTensors, from vllm.sequence import (CompletionSequenceGroupOutput, IntermediateTensors,
Logprob, SequenceData, SequenceGroupMetadata, Logprob, SequenceData, SequenceGroupMetadata,
@ -545,10 +544,6 @@ class HPUModelRunnerBase(ModelRunnerBase[TModelInputForHPU]):
] ]
gc.set_threshold(*requested_gc_thrs) gc.set_threshold(*requested_gc_thrs)
# Multi-modal data support
self.multi_modal_input_mapper = MULTIMODAL_REGISTRY \
.create_input_mapper(self.model_config)
self.skip_warmup = os.environ.get('VLLM_SKIP_WARMUP', self.skip_warmup = os.environ.get('VLLM_SKIP_WARMUP',
'false').lower() == 'true' 'false').lower() == 'true'
@ -731,9 +726,8 @@ class HPUModelRunnerBase(ModelRunnerBase[TModelInputForHPU]):
# is always the first token in the sequence. # is always the first token in the sequence.
input_positions.append(list(range(context_len, seq_len))) input_positions.append(list(range(context_len, seq_len)))
mm_data = seq_group_metadata.multi_modal_data mm_kwargs = seq_group_metadata.multi_modal_data
if mm_data: if mm_kwargs:
mm_kwargs = self.multi_modal_input_mapper(mm_data)
multi_modal_kwargs_list.append(mm_kwargs) multi_modal_kwargs_list.append(mm_kwargs)
if seq_group_metadata.block_tables is None: if seq_group_metadata.block_tables is None:

22
vllm/worker/model_runner.py
View File

@ -457,7 +457,6 @@ class ModelInputForGPUBuilder(ModelRunnerInputBuilderBase[ModelInputForGPU]):
self.enable_lora = self.runner.lora_config is not None self.enable_lora = self.runner.lora_config is not None
self.enable_prompt_adapter = (self.runner.prompt_adapter_config self.enable_prompt_adapter = (self.runner.prompt_adapter_config
is not None) is not None)
self.multi_modal_input_mapper = self.runner.multi_modal_input_mapper
# Attention metadata inputs. # Attention metadata inputs.
if self.attn_backend is not None: if self.attn_backend is not None:
@ -675,23 +674,15 @@ class ModelInputForGPUBuilder(ModelRunnerInputBuilderBase[ModelInputForGPU]):
def _compute_multi_modal_input(self, inter_data: InterDataForSeqGroup, def _compute_multi_modal_input(self, inter_data: InterDataForSeqGroup,
seq_group_metadata: SequenceGroupMetadata): seq_group_metadata: SequenceGroupMetadata):
"""If multi-modal data is given, add it to the input.""" """If multi-modal data is given, add it to the input."""
# NOTE: mm_data only includes the subset of multi-modal items that # NOTE: mm_kwargs only includes the subset of multi-modal items that
# intersect with the current prefill positions. # intersect with the current prefill positions.
positions = inter_data.input_positions[0] positions = inter_data.input_positions[0]
mm_data, placeholder_maps = MultiModalPlaceholderMap.from_seq_group( mm_kwargs, placeholder_maps = MultiModalPlaceholderMap.from_seq_group(
seq_group_metadata, seq_group_metadata,
range(positions[0], positions[0] + len(positions))) range(positions[0], positions[0] + len(positions)))
if not mm_data: if not mm_kwargs:
return return
if self.runner.mm_registry.has_processor(self.runner.model_config):
mm_kwargs = mm_data
else:
mm_kwargs = self.multi_modal_input_mapper(
mm_data,
seq_group_metadata.mm_processor_kwargs,
)
inter_data.multi_modal_kwargs = mm_kwargs inter_data.multi_modal_kwargs = mm_kwargs
inter_data.multi_modal_placeholder_maps = placeholder_maps inter_data.multi_modal_placeholder_maps = placeholder_maps
@ -1085,9 +1076,6 @@ class GPUModelRunnerBase(ModelRunnerBase[TModelInputForGPU]):
# Multi-modal data support # Multi-modal data support
self.input_registry = input_registry self.input_registry = input_registry
self.mm_registry = mm_registry self.mm_registry = mm_registry
self.multi_modal_input_mapper = mm_registry \
.create_input_mapper(model_config)
self.mm_registry.init_mm_limits_per_prompt(self.model_config)
# Lazy initialization # Lazy initialization
self.model: nn.Module # Set after load_model self.model: nn.Module # Set after load_model
@ -1327,8 +1315,8 @@ class GPUModelRunnerBase(ModelRunnerBase[TModelInputForGPU]):
dummy_data = self.input_registry \ dummy_data = self.input_registry \
.dummy_data_for_profiling(self.model_config, .dummy_data_for_profiling(self.model_config,
seq_len, seq_len,
self.mm_registry) self.mm_registry)
seq = SequenceGroupMetadata( seq = SequenceGroupMetadata(
request_id=str(group_id), request_id=str(group_id),

20
vllm/worker/neuron_model_runner.py
View File

@ -15,8 +15,7 @@ from vllm.logger import init_logger
from vllm.model_executor import SamplingMetadata from vllm.model_executor import SamplingMetadata
from vllm.model_executor.layers.sampler import SamplerOutput from vllm.model_executor.layers.sampler import SamplerOutput
from vllm.model_executor.model_loader.neuron import get_neuron_model from vllm.model_executor.model_loader.neuron import get_neuron_model
from vllm.multimodal import (MULTIMODAL_REGISTRY, BatchedTensorInputs, from vllm.multimodal import BatchedTensorInputs, MultiModalKwargs
MultiModalKwargs)
from vllm.sequence import IntermediateTensors, SequenceGroupMetadata from vllm.sequence import IntermediateTensors, SequenceGroupMetadata
from vllm.utils import is_pin_memory_available, make_tensor_with_pad from vllm.utils import is_pin_memory_available, make_tensor_with_pad
from vllm.worker.model_runner_base import ModelRunnerBase, ModelRunnerInputBase from vllm.worker.model_runner_base import ModelRunnerBase, ModelRunnerInputBase
@ -69,11 +68,6 @@ class NeuronModelRunner(ModelRunnerBase[ModelInputForNeuron]):
self.device = self.device_config.device self.device = self.device_config.device
self.pin_memory = is_pin_memory_available() self.pin_memory = is_pin_memory_available()
# Multi-modal data support
self.mm_registry = MULTIMODAL_REGISTRY
self.multi_modal_input_mapper = self.mm_registry \
.create_input_mapper(self.model_config)
# Lazy initialization. # Lazy initialization.
self.model: nn.Module # initialize after load_model. self.model: nn.Module # initialize after load_model.
@ -149,16 +143,8 @@ class NeuronModelRunner(ModelRunnerBase[ModelInputForNeuron]):
assert len(block_table) == 1 assert len(block_table) == 1
input_block_ids.append(block_table[0]) input_block_ids.append(block_table[0])
mm_data = seq_group_metadata.multi_modal_data mm_kwargs = seq_group_metadata.multi_modal_data
if mm_data: if mm_kwargs:
if self.mm_registry.has_processor(self.model_config):
mm_kwargs = mm_data
else:
mm_kwargs = self.multi_modal_input_mapper(
mm_data,
seq_group_metadata.mm_processor_kwargs,
)
multi_modal_kwargs_list.append(mm_kwargs) multi_modal_kwargs_list.append(mm_kwargs)
max_seq_len = max(seq_lens) max_seq_len = max(seq_lens)

16
vllm/worker/xpu_model_runner.py
View File

@ -188,20 +188,11 @@ class ModelInputForXPUBuilder(ModelRunnerInputBuilderBase[ModelInputForXPU]):
input_positions.extend(list(positions_range)) input_positions.extend(list(positions_range))
if seq_group_metadata.multi_modal_data: if seq_group_metadata.multi_modal_data:
# NOTE: mm_data only includes the subset of multi-modal items # NOTE: mm_kwargs only includes the subset of multi-modal items
# that intersect with the current prefill positions. # that intersect with the current prefill positions.
mm_data, placeholder_maps = MultiModalPlaceholderMap \ mm_kwargs, placeholder_maps = MultiModalPlaceholderMap \
.from_seq_group(seq_group_metadata, positions_range) .from_seq_group(seq_group_metadata, positions_range)
if self.runner.mm_registry.has_processor(
self.runner.model_config):
mm_kwargs = mm_data
else:
mm_kwargs = self.runner.multi_modal_input_mapper(
mm_data,
seq_group_metadata.mm_processor_kwargs,
)
multi_modal_kwargs_list.append(mm_kwargs) multi_modal_kwargs_list.append(mm_kwargs)
for modality, placeholder_map in placeholder_maps.items(): for modality, placeholder_map in placeholder_maps.items():
@ -404,9 +395,6 @@ class XPUModelRunner(ModelRunnerBase[ModelInputForXPUWithSamplingMetadata]):
# Multi-modal data support # Multi-modal data support
self.input_registry = input_registry self.input_registry = input_registry
self.mm_registry = mm_registry self.mm_registry = mm_registry
self.multi_modal_input_mapper = mm_registry \
.create_input_mapper(model_config)
self.mm_registry.init_mm_limits_per_prompt(self.model_config)
# Lazy initialization. # Lazy initialization.
self.model: nn.Module # Set after init_Model self.model: nn.Module # Set after init_Model