[Attention] Move Backend enum into registry (#25893)

Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
Signed-off-by: yewentao256 <zhyanwentao@126.com>

tests/compile/test_full_graph.py

@@ -11,8 +11,8 @@ import pytest
 import torch
 
 from tests.quantization.utils import is_quant_method_supported
-from tests.v1.attention.utils import _Backend
 from vllm import LLM, SamplingParams
+from vllm.attention.backends.registry import _Backend
 from vllm.attention.selector import global_force_attn_backend_context_manager
 from vllm.config import (CompilationConfig, CompilationLevel, CUDAGraphMode,
                          PassConfig)

tests/compile/test_fusion_attn.py

@@ -8,11 +8,11 @@ import torch._dynamo
 
 from tests.compile.backend import LazyInitPass, TestBackend
 from tests.models.utils import check_outputs_equal
-from tests.v1.attention.utils import (BatchSpec, _Backend,
+from tests.v1.attention.utils import BatchSpec, create_common_attn_metadata
-                                      create_common_attn_metadata)
 from vllm import LLM, SamplingParams
 from vllm._custom_ops import cutlass_scaled_fp4_mm, scaled_fp4_quant
 from vllm.attention import Attention, AttentionMetadata
+from vllm.attention.backends.registry import _Backend
 from vllm.attention.selector import global_force_attn_backend_context_manager
 from vllm.compilation.fusion import QUANT_OPS
 from vllm.compilation.fusion_attn import ATTN_OP, AttnFusionPass

tests/kernels/attention/test_mha_attn.py

@@ -10,8 +10,9 @@ from unittest.mock import patch
 import pytest
 import torch
 
+from vllm.attention.backends.registry import _Backend
 from vllm.attention.layer import MultiHeadAttention
-from vllm.attention.selector import _Backend, _cached_get_attn_backend
+from vllm.attention.selector import _cached_get_attn_backend
 from vllm.platforms import current_platform
 from vllm.platforms.cpu import CpuPlatform
 from vllm.platforms.cuda import CudaPlatform

tests/kernels/utils.py

@@ -15,10 +15,10 @@ from torch._prims_common import TensorLikeType
 
 from tests.kernels.quant_utils import native_w8a8_block_matmul
 from vllm.attention import AttentionBackend, AttentionMetadata, AttentionType
+from vllm.attention.backends.registry import _Backend
 from vllm.model_executor.layers.activation import SiluAndMul
 from vllm.model_executor.layers.fused_moe.utils import (
     moe_kernel_quantize_input)
-from vllm.platforms.interface import _Backend
 from vllm.utils import (STR_BACKEND_ENV_VAR, STR_FLASH_ATTN_VAL,
                         STR_XFORMERS_ATTN_VAL, make_tensor_with_pad)
 

tests/v1/attention/test_attention_backends.py

@@ -8,11 +8,11 @@ import pytest
 import torch
 from torch.nn.attention.flex_attention import create_block_mask, flex_attention
 
-from tests.v1.attention.utils import (BatchSpec, _Backend,
+from tests.v1.attention.utils import (BatchSpec, create_common_attn_metadata,
-                                      create_common_attn_metadata,
                                       create_standard_kv_cache_spec,
                                       create_vllm_config,
                                       get_attention_backend)
+from vllm.attention.backends.registry import _Backend
 from vllm.config import ModelConfig
 from vllm.platforms import current_platform
 from vllm.utils import STR_DTYPE_TO_TORCH_DTYPE, cdiv, is_torch_equal_or_newer

tests/v1/attention/test_mla_backends.py

@@ -6,12 +6,12 @@ from typing import Optional, Union
 import pytest
 import torch
 
-from tests.v1.attention.utils import (BatchSpec, _Backend,
+from tests.v1.attention.utils import (BatchSpec, create_common_attn_metadata,
-                                      create_common_attn_metadata,
                                       create_standard_kv_cache_spec,
                                       create_vllm_config,
                                       get_attention_backend)
 from vllm import _custom_ops as ops
+from vllm.attention.backends.registry import _Backend
 from vllm.utils import STR_DTYPE_TO_TORCH_DTYPE, cdiv
 from vllm.v1.attention.backends.utils import CommonAttentionMetadata
 from vllm.v1.kv_cache_interface import FullAttentionSpec

tests/v1/attention/utils.py

@@ -8,10 +8,11 @@ from typing import Optional, Union
 import pytest
 import torch
 
+from vllm.attention.backends.registry import _Backend
 from vllm.config import (CacheConfig, CompilationConfig, DeviceConfig,
                          LoadConfig, ModelConfig, ModelDType, ParallelConfig,
                          SchedulerConfig, VllmConfig)
-from vllm.platforms import _Backend, current_platform
+from vllm.platforms import current_platform
 from vllm.utils import resolve_obj_by_qualname
 from vllm.v1.attention.backends.utils import CommonAttentionMetadata
 from vllm.v1.kv_cache_interface import FullAttentionSpec

tests/v1/spec_decode/test_eagle.py

@@ -8,10 +8,10 @@ import pytest
 import torch
 
 from tests.utils import get_attn_backend_list_based_on_platform
-from tests.v1.attention.utils import (BatchSpec, _Backend,
+from tests.v1.attention.utils import (BatchSpec, create_common_attn_metadata,
-                                      create_common_attn_metadata,
                                       create_standard_kv_cache_spec,
                                       get_attention_backend)
+from vllm.attention.backends.registry import _Backend
 from vllm.config import (CacheConfig, DeviceConfig, ModelConfig,
                          ParallelConfig, SchedulerConfig, SpeculativeConfig,
                          VllmConfig)

tests/v1/spec_decode/test_mtp.py

@@ -6,10 +6,10 @@ from unittest import mock
 import pytest
 import torch
 
-from tests.v1.attention.utils import (BatchSpec, _Backend,
+from tests.v1.attention.utils import (BatchSpec, create_common_attn_metadata,
-                                      create_common_attn_metadata,
                                       create_standard_kv_cache_spec,
                                       get_attention_backend)
+from vllm.attention.backends.registry import _Backend
 from vllm.config import (CacheConfig, DeviceConfig, ModelConfig,
                          ParallelConfig, SchedulerConfig, SpeculativeConfig,
                          VllmConfig)

tests/v1/spec_decode/test_tree_attention.py

@@ -6,9 +6,10 @@ from typing import Optional
 
 import torch
 
-from tests.v1.attention.utils import (_Backend, create_standard_kv_cache_spec,
+from tests.v1.attention.utils import (create_standard_kv_cache_spec,
                                       create_vllm_config,
                                       get_attention_backend)
+from vllm.attention.backends.registry import _Backend
 from vllm.config import ParallelConfig, SpeculativeConfig
 from vllm.v1.attention.backends.utils import CommonAttentionMetadata
 

vllm/attention/backends/registry.py

@@ -0,0 +1,27 @@
+# SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
+"""Attention backend registry"""
+
+import enum
+
+
+class _Backend(enum.Enum):
+    FLASH_ATTN = enum.auto()
+    TRITON_ATTN = enum.auto()
+    XFORMERS = enum.auto()
+    ROCM_FLASH = enum.auto()
+    ROCM_AITER_MLA = enum.auto()
+    ROCM_AITER_FA = enum.auto()  # used for ViT attn backend
+    TORCH_SDPA = enum.auto()
+    FLASHINFER = enum.auto()
+    FLASHINFER_MLA = enum.auto()
+    TRITON_MLA = enum.auto()
+    CUTLASS_MLA = enum.auto()
+    FLASHMLA = enum.auto()
+    FLASH_ATTN_MLA = enum.auto()
+    PALLAS = enum.auto()
+    IPEX = enum.auto()
+    NO_ATTENTION = enum.auto()
+    FLEX_ATTENTION = enum.auto()
+    TREE_ATTN = enum.auto()
+    ROCM_ATTN = enum.auto()

vllm/attention/layer.py

@@ -10,6 +10,7 @@ import torch.nn.functional as F
 import vllm.envs as envs
 from vllm.attention import AttentionType
 from vllm.attention.backends.abstract import AttentionBackend
+from vllm.attention.backends.registry import _Backend
 from vllm.attention.selector import backend_name_to_enum, get_attn_backend
 from vllm.attention.utils.kv_sharing_utils import validate_kv_sharing_target
 from vllm.config import CacheConfig, get_current_vllm_config
@@ -26,7 +27,7 @@ from vllm.model_executor.layers.quantization.kv_cache import BaseKVCacheMethod
 from vllm.model_executor.layers.quantization.utils.quant_utils import (
     GroupShape)
 from vllm.model_executor.models.vision import get_vit_attn_backend
-from vllm.platforms import _Backend, current_platform
+from vllm.platforms import current_platform
 from vllm.utils import GiB_bytes, direct_register_custom_op
 
 logger = init_logger(__name__)

vllm/attention/selector.py

@@ -11,8 +11,9 @@ import torch
 
 import vllm.envs as envs
 from vllm.attention.backends.abstract import AttentionBackend
+from vllm.attention.backends.registry import _Backend
 from vllm.logger import init_logger
-from vllm.platforms import _Backend, current_platform
+from vllm.platforms import current_platform
 from vllm.utils import STR_BACKEND_ENV_VAR, resolve_obj_by_qualname
 
 logger = init_logger(__name__)

vllm/distributed/kv_transfer/kv_connector/v1/nixl_connector.py

@@ -20,6 +20,7 @@ import torch
 import zmq
 
 from vllm import envs
+from vllm.attention.backends.registry import _Backend
 from vllm.attention.selector import backend_name_to_enum, get_attn_backend
 from vllm.config import VllmConfig
 from vllm.distributed.kv_transfer.kv_connector.v1.base import (
@@ -32,7 +33,7 @@ from vllm.distributed.parallel_state import (
 from vllm.distributed.utils import divide
 from vllm.forward_context import ForwardContext
 from vllm.logger import init_logger
-from vllm.platforms import _Backend, current_platform
+from vllm.platforms import current_platform
 from vllm.utils import make_zmq_path, make_zmq_socket
 from vllm.v1.attention.backends.utils import get_kv_cache_layout
 from vllm.v1.core.sched.output import SchedulerOutput

vllm/envs.py

@@ -619,8 +619,9 @@ environment_variables: dict[str, Callable[[], Any]] = {
     # All possible options loaded dynamically from _Backend enum
     "VLLM_ATTENTION_BACKEND":
     env_with_choices("VLLM_ATTENTION_BACKEND", None,
-                     lambda: list(__import__('vllm.platforms.interface', \
+                     lambda: list(__import__(
-                        fromlist=['_Backend'])._Backend.__members__.keys())),
+                         'vllm.attention.backends.registry',
+                         fromlist=['_Backend'])._Backend.__members__.keys())),
 
     # If set, vllm will use flashinfer sampler
     "VLLM_USE_FLASHINFER_SAMPLER":

vllm/model_executor/models/dots_ocr.py

@@ -9,6 +9,7 @@ import torch.nn.functional as F
 from torch.nn import LayerNorm
 from transformers.models.qwen2_vl import Qwen2VLProcessor
 
+from vllm.attention.backends.registry import _Backend
 from vllm.attention.layer import check_upstream_fa_availability
 from vllm.config import VllmConfig
 from vllm.distributed import utils as dist_utils
@@ -38,7 +39,6 @@ from vllm.model_executor.models.utils import (AutoWeightsLoader, WeightsMapper,
 from vllm.model_executor.models.vision import get_vit_attn_backend
 from vllm.multimodal import MULTIMODAL_REGISTRY
 from vllm.multimodal.inputs import MultiModalDataDict
-from vllm.platforms import _Backend
 from vllm.sequence import IntermediateTensors
 from vllm.transformers_utils.configs.dotsocr import (DotsOCRConfig,
                                                      DotsVisionConfig)

vllm/model_executor/models/ernie45_vl.py

@@ -34,6 +34,7 @@ import torch.nn.functional as F
 from einops import rearrange, repeat
 from transformers import BatchFeature
 
+from vllm.attention.backends.registry import _Backend
 from vllm.attention.layer import check_upstream_fa_availability
 from vllm.config import VllmConfig
 from vllm.distributed import parallel_state
@@ -54,7 +55,7 @@ from vllm.multimodal.processing import (BaseMultiModalProcessor,
                                         BaseProcessingInfo, PromptReplacement,
                                         PromptUpdate)
 from vllm.multimodal.profiling import BaseDummyInputsBuilder
-from vllm.platforms import _Backend, current_platform
+from vllm.platforms import current_platform
 from vllm.sequence import IntermediateTensors
 from vllm.utils.tensor_schema import TensorSchema, TensorShape
 

vllm/model_executor/models/glm4_1v.py

@@ -46,6 +46,7 @@ from transformers.models.glm4v.video_processing_glm4v import (
     Glm4vVideoProcessor)
 from transformers.video_utils import VideoMetadata
 
+from vllm.attention.backends.registry import _Backend
 from vllm.attention.layer import check_upstream_fa_availability
 from vllm.config import VllmConfig
 from vllm.distributed import (get_tensor_model_parallel_world_size,
@@ -69,7 +70,6 @@ from vllm.multimodal.processing import (BaseMultiModalProcessor,
                                         BaseProcessingInfo, PromptReplacement,
                                         PromptUpdate, PromptUpdateDetails)
 from vllm.multimodal.profiling import BaseDummyInputsBuilder
-from vllm.platforms import _Backend
 from vllm.sequence import IntermediateTensors
 from vllm.utils.tensor_schema import TensorSchema, TensorShape
 

vllm/model_executor/models/keye.py

@@ -17,6 +17,7 @@ from transformers.modeling_outputs import (BaseModelOutput,
                                            BaseModelOutputWithPooling)
 from transformers.utils import torch_int
 
+from vllm.attention.backends.registry import _Backend
 from vllm.attention.layer import check_upstream_fa_availability
 from vllm.config import VllmConfig
 from vllm.distributed import get_tensor_model_parallel_world_size
@@ -39,7 +40,6 @@ from vllm.multimodal.processing import (BaseMultiModalProcessor,
                                         BaseProcessingInfo, PromptReplacement,
                                         PromptUpdate)
 from vllm.multimodal.profiling import BaseDummyInputsBuilder
-from vllm.platforms import _Backend
 from vllm.sequence import IntermediateTensors
 from vllm.utils import is_list_of
 from vllm.utils.tensor_schema import TensorSchema, TensorShape

vllm/model_executor/models/qwen2_5_vl.py

@@ -38,6 +38,7 @@ from transformers.models.qwen2_5_vl import Qwen2_5_VLProcessor
 from transformers.models.qwen2_5_vl.configuration_qwen2_5_vl import (
     Qwen2_5_VLConfig, Qwen2_5_VLVisionConfig)
 
+from vllm.attention.backends.registry import _Backend
 from vllm.attention.layer import check_upstream_fa_availability
 from vllm.config import VllmConfig
 from vllm.distributed import parallel_state
@@ -62,7 +63,6 @@ from vllm.multimodal.evs import (compute_mrope_for_media,
 from vllm.multimodal.inputs import MultiModalFieldConfig, MultiModalKwargs
 from vllm.multimodal.parse import MultiModalDataItems
 from vllm.multimodal.processing import PromptReplacement, PromptUpdate
-from vllm.platforms import _Backend
 from vllm.sequence import IntermediateTensors
 from vllm.utils import is_pin_memory_available
 from vllm.utils.tensor_schema import TensorSchema, TensorShape

vllm/model_executor/models/qwen2_vl.py

@@ -41,6 +41,7 @@ from transformers.models.qwen2_vl.image_processing_qwen2_vl import smart_resize
 from transformers.models.qwen2_vl.video_processing_qwen2_vl import (
     Qwen2VLVideoProcessor)
 
+from vllm.attention.backends.registry import _Backend
 from vllm.attention.layer import check_upstream_fa_availability
 from vllm.config import VllmConfig
 from vllm.distributed import parallel_state, tensor_model_parallel_all_gather
@@ -65,7 +66,6 @@ from vllm.multimodal.processing import (BaseMultiModalProcessor,
                                         BaseProcessingInfo, PromptReplacement,
                                         PromptUpdate)
 from vllm.multimodal.profiling import BaseDummyInputsBuilder
-from vllm.platforms import _Backend
 from vllm.sequence import IntermediateTensors
 from vllm.transformers_utils.tokenizer import AnyTokenizer
 from vllm.utils.tensor_schema import TensorSchema, TensorShape

vllm/model_executor/models/qwen3_vl.py

@@ -43,6 +43,7 @@ from transformers.models.qwen3_vl.video_processing_qwen3_vl import (
     smart_resize as video_smart_resize)
 from transformers.video_utils import VideoMetadata
 
+from vllm.attention.backends.registry import _Backend
 from vllm.attention.layer import check_upstream_fa_availability
 from vllm.compilation.decorators import support_torch_compile
 from vllm.config import VllmConfig
@@ -66,7 +67,6 @@ from vllm.multimodal.processing import (BaseMultiModalProcessor,
                                         PromptReplacement, PromptUpdate,
                                         PromptUpdateDetails)
 from vllm.multimodal.profiling import BaseDummyInputsBuilder
-from vllm.platforms import _Backend
 from vllm.sequence import IntermediateTensors
 from vllm.utils import is_list_of
 

vllm/model_executor/models/siglip2navit.py

@@ -13,6 +13,7 @@ from torch.nn import functional as F
 from transformers import Siglip2VisionConfig
 from transformers.configuration_utils import PretrainedConfig
 
+from vllm.attention.backends.registry import _Backend
 from vllm.attention.layer import check_upstream_fa_availability
 from vllm.distributed import divide, get_tensor_model_parallel_world_size
 from vllm.model_executor.layers.activation import get_act_fn
@@ -22,7 +23,6 @@ from vllm.model_executor.layers.linear import (ColumnParallelLinear,
                                                RowParallelLinear)
 from vllm.model_executor.layers.quantization import QuantizationConfig
 from vllm.model_executor.model_loader.weight_utils import default_weight_loader
-from vllm.platforms import _Backend
 
 from .vision import get_vit_attn_backend
 

vllm/model_executor/models/vision.py

@@ -10,11 +10,12 @@ from typing import (Callable, Final, Generic, Literal, Optional, Protocol,
 import torch
 from transformers import PretrainedConfig
 
+from vllm.attention.backends.registry import _Backend
 from vllm.distributed import (get_tensor_model_parallel_rank,
                               get_tensor_model_parallel_world_size,
                               tensor_model_parallel_all_gather)
 from vllm.logger import init_logger
-from vllm.platforms import _Backend, current_platform
+from vllm.platforms import current_platform
 
 logger = init_logger(__name__)
 

vllm/platforms/__init__.py

@@ -9,7 +9,6 @@ from vllm import envs
 from vllm.plugins import load_plugins_by_group
 from vllm.utils import resolve_obj_by_qualname, supports_xccl
 
-from .interface import _Backend  # noqa: F401
 from .interface import CpuArchEnum, Platform, PlatformEnum
 
 logger = logging.getLogger(__name__)

vllm/platforms/cpu.py

@@ -15,13 +15,15 @@ import torch
 from vllm.logger import init_logger
 from vllm.utils import DEFAULT_MAX_NUM_BATCHED_TOKENS
 
-from .interface import CpuArchEnum, Platform, PlatformEnum, _Backend
+from .interface import CpuArchEnum, Platform, PlatformEnum
 
 logger = init_logger(__name__)
 
 if TYPE_CHECKING:
+    from vllm.attention.backends.registry import _Backend
     from vllm.config import VllmConfig
 else:
+    _Backend = None
     VllmConfig = None
 
 
@@ -90,10 +92,11 @@ class CpuPlatform(Platform):
         return "cpu"
 
     @classmethod
-    def get_attn_backend_cls(cls, selected_backend: _Backend, head_size: int,
+    def get_attn_backend_cls(cls, selected_backend: "_Backend", head_size: int,
                              dtype: torch.dtype, kv_cache_dtype: Optional[str],
                              block_size: int, use_v1: bool, use_mla: bool,
                              has_sink: bool, use_sparse: bool) -> str:
+        from vllm.attention.backends.registry import _Backend
         if selected_backend and selected_backend != _Backend.TORCH_SDPA:
             logger.info("Cannot use %s backend on CPU.", selected_backend)
         if use_mla:

vllm/platforms/cuda.py

@@ -20,10 +20,13 @@ import vllm.envs as envs
 from vllm.logger import init_logger
 from vllm.utils import cuda_device_count_stateless, import_pynvml
 
-from .interface import DeviceCapability, Platform, PlatformEnum, _Backend
+from .interface import DeviceCapability, Platform, PlatformEnum
 
 if TYPE_CHECKING:
+    from vllm.attention.backends.registry import _Backend
     from vllm.config import ModelConfig, VllmConfig
+else:
+    _Backend = None
 
 logger = init_logger(__name__)
 
@@ -202,7 +205,8 @@ class CudaPlatformBase(Platform):
 
     @classmethod
     def get_vit_attn_backend(cls, head_size: int,
-                             dtype: torch.dtype) -> _Backend:
+                             dtype: torch.dtype) -> "_Backend":
+        from vllm.attention.backends.registry import _Backend
 
         # For Blackwell GPUs, force TORCH_SDPA for now.
         # See https://github.com/facebookresearch/xformers/issues/1317#issuecomment-3199392579 # noqa: E501
@@ -230,6 +234,7 @@ class CudaPlatformBase(Platform):
     def get_attn_backend_cls(cls, selected_backend, head_size, dtype,
                              kv_cache_dtype, block_size, use_v1, use_mla,
                              has_sink, use_sparse) -> str:
+        from vllm.attention.backends.registry import _Backend
         if use_mla:
             if not use_v1:
                 raise RuntimeError(

vllm/platforms/interface.py

@@ -17,12 +17,14 @@ from vllm.inputs import ProcessorInputs, PromptType
 from vllm.logger import init_logger
 
 if TYPE_CHECKING:
+    from vllm.attention.backends.registry import _Backend
     from vllm.config import ModelConfig, VllmConfig
     from vllm.lora.request import LoRARequest
     from vllm.pooling_params import PoolingParams
     from vllm.sampling_params import SamplingParams
     from vllm.utils import FlexibleArgumentParser
 else:
+    _Backend = None
     ModelConfig = None
     VllmConfig = None
     LoRARequest = None
@@ -38,30 +40,6 @@ def in_wsl() -> bool:
     return "microsoft" in " ".join(uname()).lower()
 
 
-class _Backend(enum.Enum):
-    FLASH_ATTN = enum.auto()
-    TRITON_ATTN = enum.auto()
-    XFORMERS = enum.auto()
-    ROCM_FLASH = enum.auto()
-    ROCM_AITER_MLA = enum.auto()  # Supported by V1
-    ROCM_AITER_FA = enum.auto()  # used for ViT attn backend
-    TORCH_SDPA = enum.auto()
-    FLASHINFER = enum.auto()
-    FLASHINFER_MLA = enum.auto()
-    TRITON_MLA = enum.auto()  # Supported by V1
-    CUTLASS_MLA = enum.auto()
-    FLASHMLA = enum.auto()  # Supported by V1
-    FLASH_ATTN_MLA = enum.auto()  # Supported by V1
-    PALLAS = enum.auto()
-    IPEX = enum.auto()
-    DUAL_CHUNK_FLASH_ATTN = enum.auto()
-    DIFFERENTIAL_FLASH_ATTN = enum.auto()
-    NO_ATTENTION = enum.auto()
-    FLEX_ATTENTION = enum.auto()
-    TREE_ATTN = enum.auto()
-    ROCM_ATTN = enum.auto()
-
-
 class PlatformEnum(enum.Enum):
     CUDA = enum.auto()
     ROCM = enum.auto()
@@ -187,11 +165,12 @@ class Platform:
 
     @classmethod
     def get_vit_attn_backend(cls, head_size: int,
-                             dtype: torch.dtype) -> _Backend:
+                             dtype: torch.dtype) -> "_Backend":
+        from vllm.attention.backends.registry import _Backend
         return _Backend.TORCH_SDPA
 
     @classmethod
-    def get_attn_backend_cls(cls, selected_backend: _Backend, head_size: int,
+    def get_attn_backend_cls(cls, selected_backend: "_Backend", head_size: int,
                              dtype: torch.dtype, kv_cache_dtype: Optional[str],
                              block_size: int, use_v1: bool, use_mla: bool,
                              has_sink: bool, use_sparse: bool) -> str:

vllm/platforms/rocm.py

@@ -14,10 +14,13 @@ import vllm.envs as envs
 from vllm.logger import init_logger
 from vllm.utils import cuda_device_count_stateless
 
-from .interface import DeviceCapability, Platform, PlatformEnum, _Backend
+from .interface import DeviceCapability, Platform, PlatformEnum
 
 if TYPE_CHECKING:
+    from vllm.attention.backends.registry import _Backend
     from vllm.config import ModelConfig, VllmConfig
+else:
+    _Backend = None
 
 logger = init_logger(__name__)
 
@@ -182,7 +185,8 @@ class RocmPlatform(Platform):
 
     @classmethod
     def get_vit_attn_backend(cls, head_size: int,
-                             dtype: torch.dtype) -> _Backend:
+                             dtype: torch.dtype) -> "_Backend":
+        from vllm.attention.backends.registry import _Backend
         if (envs.VLLM_ROCM_USE_AITER and envs.VLLM_ROCM_USE_AITER_MHA
                 and on_gfx9()):
             # Note: AITER FA is only supported for Qwen-VL models.
@@ -196,6 +200,7 @@ class RocmPlatform(Platform):
     def get_attn_backend_cls(cls, selected_backend, head_size, dtype,
                              kv_cache_dtype, block_size, use_v1, use_mla,
                              has_sink, use_sparse) -> str:
+        from vllm.attention.backends.registry import _Backend
         if use_sparse:
             raise NotImplementedError(
                 "Sparse Attention is not supported on ROCm.")

vllm/platforms/tpu.py

@@ -11,9 +11,10 @@ from vllm.logger import init_logger
 from vllm.sampling_params import SamplingParams, SamplingType
 from vllm.utils import DEFAULT_MAX_NUM_BATCHED_TOKENS
 
-from .interface import Platform, PlatformEnum, _Backend
+from .interface import Platform, PlatformEnum
 
 if TYPE_CHECKING:
+    from vllm.attention.backends.registry import _Backend
     from vllm.config import BlockSize, ModelConfig, VllmConfig
     from vllm.pooling_params import PoolingParams
 else:
@@ -21,6 +22,7 @@ else:
     ModelConfig = None
     VllmConfig = None
     PoolingParams = None
+    _Backend = None
 
 logger = init_logger(__name__)
 
@@ -46,10 +48,11 @@ class TpuPlatform(Platform):
     ]
 
     @classmethod
-    def get_attn_backend_cls(cls, selected_backend: _Backend, head_size: int,
+    def get_attn_backend_cls(cls, selected_backend: "_Backend", head_size: int,
                              dtype: torch.dtype, kv_cache_dtype: Optional[str],
                              block_size: int, use_v1: bool, use_mla: bool,
                              has_sink, use_sparse) -> str:
+        from vllm.attention.backends.registry import _Backend
         if use_sparse:
             raise NotImplementedError(
                 "Sparse Attention is not supported on TPU.")

vllm/platforms/xpu.py

@@ -10,13 +10,15 @@ import vllm.envs as envs
 from vllm.logger import init_logger
 from vllm.utils import DEFAULT_MAX_NUM_BATCHED_TOKENS
 
-from .interface import DeviceCapability, Platform, PlatformEnum, _Backend
+from .interface import DeviceCapability, Platform, PlatformEnum
 
 if TYPE_CHECKING:
+    from vllm.attention.backends.registry import _Backend
     from vllm.config import ModelConfig, VllmConfig
 else:
     ModelConfig = None
     VllmConfig = None
+    _Backend = None
 
 logger = init_logger(__name__)
 
@@ -33,10 +35,11 @@ class XPUPlatform(Platform):
     device_control_env_var: str = "ZE_AFFINITY_MASK"
 
     @classmethod
-    def get_attn_backend_cls(cls, selected_backend: _Backend, head_size: int,
+    def get_attn_backend_cls(cls, selected_backend: "_Backend", head_size: int,
                              dtype: torch.dtype, kv_cache_dtype: Optional[str],
                              block_size: int, use_v1: bool, use_mla: bool,
                              has_sink: bool, use_sparse) -> str:
+        from vllm.attention.backends.registry import _Backend
         if use_sparse:
             raise NotImplementedError(
                 "Sparse Attention is not supported on XPU.")