Skip to content

vllm.model_executor.models.transformers

Wrapper around transformers models

Modules:

Name Description
base

Transformers modeling backend base class.
causal

Transformers modeling backend mixin for causal language models.
legacy

Transformers modeling backend mixin for legacy models.
moe

Transformers modeling backend mixin for Mixture of Experts (MoE) models.
multimodal

Transformers modeling backend mixin for multi-modal models.
pooling

Transformers modeling backend mixins for pooling models.
utils

Transformers modeling backend utilities.

TransformersEmbeddingModel

Bases: EmbeddingMixin, LegacyMixin, Base

Source code in vllm/model_executor/models/transformers/__init__.py 
74
75
@support_torch_compile(enable_if=can_enable_torch_compile)
class TransformersEmbeddingModel(EmbeddingMixin, LegacyMixin, Base): ...

TransformersForCausalLM

Bases: CausalMixin, Base

Source code in vllm/model_executor/models/transformers/__init__.py 
40
41
@support_torch_compile(enable_if=can_enable_torch_compile)
class TransformersForCausalLM(CausalMixin, Base): ...

TransformersForSequenceClassification

Bases: SequenceClassificationMixin, LegacyMixin, Base

Source code in vllm/model_executor/models/transformers/__init__.py 
94
95
96
97
@support_torch_compile(enable_if=can_enable_torch_compile)
class TransformersForSequenceClassification(
    SequenceClassificationMixin, LegacyMixin, Base
): ...

TransformersMoEEmbeddingModel

Bases: EmbeddingMixin, MoEMixin, Base

Source code in vllm/model_executor/models/transformers/__init__.py 
78
79
@support_torch_compile(enable_if=can_enable_torch_compile)
class TransformersMoEEmbeddingModel(EmbeddingMixin, MoEMixin, Base): ...

TransformersMoEForCausalLM

Bases: MoEMixin, CausalMixin, Base

Source code in vllm/model_executor/models/transformers/__init__.py 
44
45
@support_torch_compile(enable_if=can_enable_torch_compile)
class TransformersMoEForCausalLM(MoEMixin, CausalMixin, Base): ...

TransformersMoEForSequenceClassification

Bases: SequenceClassificationMixin, MoEMixin, Base

Source code in vllm/model_executor/models/transformers/__init__.py 
100
101
102
103
@support_torch_compile(enable_if=can_enable_torch_compile)
class TransformersMoEForSequenceClassification(
    SequenceClassificationMixin, MoEMixin, Base
): ...

TransformersMultiModalEmbeddingModel

Bases: EmbeddingMixin, MultiModalMixin, Base

Source code in vllm/model_executor/models/transformers/__init__.py 
82
83
84
85
86
87
88
89
90
@MULTIMODAL_REGISTRY.register_processor(
    MultiModalProcessor,
    info=MultiModalProcessingInfo,
    dummy_inputs=MultiModalDummyInputsBuilder,
)
@support_torch_compile(
    dynamic_arg_dims=DYNAMIC_ARG_DIMS, enable_if=can_enable_torch_compile
)
class TransformersMultiModalEmbeddingModel(EmbeddingMixin, MultiModalMixin, Base): ...

TransformersMultiModalForCausalLM

Bases: MultiModalMixin, CausalMixin, Base

Source code in vllm/model_executor/models/transformers/__init__.py 
49
50
51
52
53
54
55
56
57
@MULTIMODAL_REGISTRY.register_processor(
    MultiModalProcessor,
    info=MultiModalProcessingInfo,
    dummy_inputs=MultiModalDummyInputsBuilder,
)
@support_torch_compile(
    dynamic_arg_dims=DYNAMIC_ARG_DIMS, enable_if=can_enable_torch_compile
)
class TransformersMultiModalForCausalLM(MultiModalMixin, CausalMixin, Base): ...

TransformersMultiModalForSequenceClassification

Bases: SequenceClassificationMixin, MultiModalMixin, Base

Source code in vllm/model_executor/models/transformers/__init__.py 
106
107
108
109
110
111
112
113
114
115
116
@MULTIMODAL_REGISTRY.register_processor(
    MultiModalProcessor,
    info=MultiModalProcessingInfo,
    dummy_inputs=MultiModalDummyInputsBuilder,
)
@support_torch_compile(
    dynamic_arg_dims=DYNAMIC_ARG_DIMS, enable_if=can_enable_torch_compile
)
class TransformersMultiModalForSequenceClassification(
    SequenceClassificationMixin, MultiModalMixin, Base
): ...

TransformersMultiModalMoEForCausalLM

Bases: MoEMixin, MultiModalMixin, CausalMixin, Base

Source code in vllm/model_executor/models/transformers/__init__.py 
60
61
62
63
64
65
66
67
68
69
70
@MULTIMODAL_REGISTRY.register_processor(
    MultiModalProcessor,
    info=MultiModalProcessingInfo,
    dummy_inputs=MultiModalDummyInputsBuilder,
)
@support_torch_compile(
    dynamic_arg_dims=DYNAMIC_ARG_DIMS, enable_if=can_enable_torch_compile
)
class TransformersMultiModalMoEForCausalLM(
    MoEMixin, MultiModalMixin, CausalMixin, Base
): ...

__getattr__ 

__getattr__(name: str)

Handle imports of non-existent classes with a helpful error message.

Source code in vllm/model_executor/models/transformers/__init__.py 
119
120
121
122
123
124
125
126
127
def __getattr__(name: str):
    """Handle imports of non-existent classes with a helpful error message."""
    if name not in globals():
        raise AttributeError(
            "The Transformers modeling backend does not currently have a class to "
            f"handle the requested model type: {name}. Please open an issue at "
            "https://github.com/vllm-project/vllm/issues/new"
        )
    return globals()[name]