[5/N][Attention] Finish eliminating `vllm/attention` folder (#32064)

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

vllm/model_executor/models/exaone4.py

@@ -28,11 +28,11 @@ import torch
 from torch import nn
 from transformers import Exaone4Config
 
-from vllm.attention.layer import Attention
 from vllm.compilation.decorators import support_torch_compile
 from vllm.config import CacheConfig, VllmConfig
 from vllm.distributed import get_pp_group, get_tensor_model_parallel_world_size
 from vllm.model_executor.layers.activation import SiluAndMul
+from vllm.model_executor.layers.attention import Attention
 from vllm.model_executor.layers.layernorm import RMSNorm
 from vllm.model_executor.layers.linear import (
     MergedColumnParallelLinear,

vllm/model_executor/models/falcon.py

@@ -30,7 +30,6 @@ from torch import nn
 from torch.nn import LayerNorm
 from transformers import FalconConfig as HF_FalconConfig
 
-from vllm.attention.layer import Attention
 from vllm.compilation.decorators import support_torch_compile
 from vllm.config import CacheConfig, VllmConfig
 from vllm.distributed import (
@@ -40,6 +39,7 @@ from vllm.distributed import (
     tensor_model_parallel_all_reduce,
 )
 from vllm.model_executor.layers.activation import get_act_fn
+from vllm.model_executor.layers.attention import Attention
 from vllm.model_executor.layers.linear import (
     ColumnParallelLinear,
     QKVParallelLinear,

vllm/model_executor/models/falcon_h1.py

@@ -9,12 +9,12 @@ import torch
 from torch import nn
 from transformers import FalconH1Config
 
-from vllm.attention.layer import Attention
 from vllm.compilation.decorators import support_torch_compile
 from vllm.config import CacheConfig, ModelConfig, VllmConfig
 from vllm.distributed import get_tensor_model_parallel_world_size
 from vllm.distributed.parallel_state import get_pp_group
 from vllm.model_executor.layers.activation import SiluAndMul
+from vllm.model_executor.layers.attention import Attention
 from vllm.model_executor.layers.layernorm import RMSNorm
 from vllm.model_executor.layers.linear import (
     MergedColumnParallelLinear,

vllm/model_executor/models/gemma.py

@@ -26,12 +26,12 @@ import torch
 from torch import nn
 from transformers import GemmaConfig
 
-from vllm.attention.layer import Attention
 from vllm.compilation.decorators import support_torch_compile
 from vllm.config import CacheConfig, VllmConfig
 from vllm.distributed import get_pp_group, get_tensor_model_parallel_world_size
 from vllm.logger import init_logger
 from vllm.model_executor.layers.activation import GeluAndMul
+from vllm.model_executor.layers.attention import Attention
 from vllm.model_executor.layers.layernorm import GemmaRMSNorm
 from vllm.model_executor.layers.linear import (
     MergedColumnParallelLinear,

vllm/model_executor/models/gemma2.py

@@ -23,12 +23,12 @@ import torch
 from torch import nn
 from transformers import Gemma2Config
 
-from vllm.attention.layer import Attention
 from vllm.compilation.decorators import support_torch_compile
 from vllm.config import CacheConfig, VllmConfig
 from vllm.distributed import get_pp_group, get_tensor_model_parallel_world_size
 from vllm.logger import init_logger
 from vllm.model_executor.layers.activation import GeluAndMul
+from vllm.model_executor.layers.attention import Attention
 from vllm.model_executor.layers.layernorm import GemmaRMSNorm
 from vllm.model_executor.layers.linear import (
     MergedColumnParallelLinear,

vllm/model_executor/models/gemma3.py

@@ -22,13 +22,13 @@ import torch
 from torch import nn
 from transformers import Gemma3TextConfig
 
-from vllm.attention.layer import Attention
 from vllm.compilation.decorators import support_torch_compile
 from vllm.config import CacheConfig, VllmConfig
 from vllm.distributed import get_pp_group, get_tensor_model_parallel_world_size
 from vllm.logger import init_logger
 from vllm.model_executor.layers.activation import GeluAndMul
-from vllm.model_executor.layers.attention.encoder_only_attention import (
+from vllm.model_executor.layers.attention import (
+    Attention,
     EncoderOnlyAttention,
 )
 from vllm.model_executor.layers.layernorm import GemmaRMSNorm

vllm/model_executor/models/gemma3n.py

@@ -21,7 +21,6 @@ import torch
 from torch import nn
 from transformers.models.gemma3n.configuration_gemma3n import Gemma3nTextConfig
 
-from vllm.attention.layer import Attention
 from vllm.compilation.decorators import support_torch_compile
 from vllm.config import CacheConfig, VllmConfig
 from vllm.distributed import get_tensor_model_parallel_world_size
@@ -32,6 +31,7 @@ from vllm.model_executor.layers.activation import (
     GeluAndMul,
     GeluAndMulSparse,
 )
+from vllm.model_executor.layers.attention import Attention
 from vllm.model_executor.layers.layernorm import RMSNorm
 from vllm.model_executor.layers.linear import (
     ColumnParallelLinear,

vllm/model_executor/models/glm4.py

@@ -29,10 +29,10 @@ import torch
 from torch import nn
 from transformers import Glm4Config
 
-from vllm.attention.layer import Attention
 from vllm.compilation.decorators import support_torch_compile
 from vllm.config import CacheConfig, VllmConfig
 from vllm.distributed import get_pp_group, get_tensor_model_parallel_world_size
+from vllm.model_executor.layers.attention import Attention
 from vllm.model_executor.layers.layernorm import RMSNorm
 from vllm.model_executor.layers.linear import QKVParallelLinear, RowParallelLinear
 from vllm.model_executor.layers.logits_processor import LogitsProcessor

vllm/model_executor/models/glm4_1v.py

@@ -52,7 +52,7 @@ from vllm.config.multimodal import BaseDummyOptions, VideoDummyOptions
 from vllm.distributed import get_tensor_model_parallel_world_size, parallel_state
 from vllm.distributed import utils as dist_utils
 from vllm.logger import init_logger
-from vllm.model_executor.layers.attention.mm_encoder_attention import (
+from vllm.model_executor.layers.attention import (
     MMEncoderAttention,
 )
 from vllm.model_executor.layers.conv import Conv2dLayer, Conv3dLayer

vllm/model_executor/models/glm4_moe.py

@@ -32,7 +32,6 @@ import torch
 from torch import nn
 from transformers.models.glm4_moe import Glm4MoeConfig
 
-from vllm.attention.layer import Attention
 from vllm.compilation.decorators import support_torch_compile
 from vllm.config import CacheConfig, VllmConfig, get_current_vllm_config
 from vllm.distributed import (
@@ -42,6 +41,7 @@ from vllm.distributed import (
 )
 from vllm.logger import init_logger
 from vllm.model_executor.layers.activation import SiluAndMul
+from vllm.model_executor.layers.attention import Attention
 from vllm.model_executor.layers.fused_moe import SharedFusedMoE
 from vllm.model_executor.layers.layernorm import RMSNorm
 from vllm.model_executor.layers.linear import (

vllm/model_executor/models/glm4v.py

@@ -23,7 +23,7 @@ from vllm.config import VllmConfig
 from vllm.config.multimodal import BaseDummyOptions
 from vllm.distributed import get_tensor_model_parallel_world_size
 from vllm.model_executor.layers.activation import SiluAndMul, get_act_fn
-from vllm.model_executor.layers.attention.mm_encoder_attention import MMEncoderAttention
+from vllm.model_executor.layers.attention import MMEncoderAttention
 from vllm.model_executor.layers.conv import Conv2dLayer
 from vllm.model_executor.layers.linear import (
     ColumnParallelLinear,

vllm/model_executor/models/glmasr.py

@@ -16,7 +16,7 @@ from vllm.config.multimodal import BaseDummyOptions
 from vllm.distributed.parallel_state import get_tensor_model_parallel_world_size
 from vllm.inputs.data import PromptType
 from vllm.model_executor.layers.activation import get_act_fn
-from vllm.model_executor.layers.attention.mm_encoder_attention import MMEncoderAttention
+from vllm.model_executor.layers.attention import MMEncoderAttention
 from vllm.model_executor.layers.linear import (
     ColumnParallelLinear,
     QKVParallelLinear,

vllm/model_executor/models/gpt2.py

@@ -27,7 +27,6 @@ import torch
 from torch import nn
 from transformers import GPT2Config
 
-from vllm.attention.layer import Attention
 from vllm.compilation.decorators import support_torch_compile
 from vllm.config import CacheConfig, VllmConfig
 from vllm.distributed.parallel_state import (
@@ -35,6 +34,7 @@ from vllm.distributed.parallel_state import (
     get_tensor_model_parallel_world_size,
 )
 from vllm.model_executor.layers.activation import get_act_fn
+from vllm.model_executor.layers.attention import Attention
 from vllm.model_executor.layers.linear import (
     ColumnParallelLinear,
     QKVParallelLinear,

vllm/model_executor/models/gpt_bigcode.py

@@ -28,11 +28,11 @@ import torch
 from torch import nn
 from transformers import GPTBigCodeConfig
 
-from vllm.attention.layer import Attention
 from vllm.compilation.decorators import support_torch_compile
 from vllm.config import CacheConfig, VllmConfig
 from vllm.distributed import get_pp_group, get_tensor_model_parallel_world_size
 from vllm.model_executor.layers.activation import get_act_fn
+from vllm.model_executor.layers.attention import Attention
 from vllm.model_executor.layers.linear import (
     ColumnParallelLinear,
     QKVParallelLinear,

vllm/model_executor/models/gpt_j.py

@@ -26,11 +26,11 @@ import torch
 from torch import nn
 from transformers import GPTJConfig
 
-from vllm.attention.layer import Attention
 from vllm.compilation.decorators import support_torch_compile
 from vllm.config import CacheConfig, VllmConfig
 from vllm.distributed import get_pp_group, get_tensor_model_parallel_world_size
 from vllm.model_executor.layers.activation import get_act_fn
+from vllm.model_executor.layers.attention import Attention
 from vllm.model_executor.layers.linear import (
     ColumnParallelLinear,
     QKVParallelLinear,

vllm/model_executor/models/gpt_neox.py

@@ -26,11 +26,11 @@ import torch
 from torch import nn
 from transformers import GPTNeoXConfig
 
-from vllm.attention.layer import Attention
 from vllm.compilation.decorators import support_torch_compile
 from vllm.config import CacheConfig, VllmConfig
 from vllm.distributed import get_pp_group, get_tensor_model_parallel_world_size
 from vllm.model_executor.layers.activation import get_act_fn
+from vllm.model_executor.layers.attention import Attention
 from vllm.model_executor.layers.linear import (
     ColumnParallelLinear,
     QKVParallelLinear,

vllm/model_executor/models/gpt_oss.py

@@ -7,7 +7,6 @@ import torch.distributed as dist
 from torch import nn
 from transformers import GptOssConfig
 
-from vllm.attention.layer import Attention
 from vllm.compilation.decorators import support_torch_compile
 from vllm.config import CacheConfig, VllmConfig
 from vllm.distributed import (
@@ -19,6 +18,7 @@ from vllm.distributed import (
     get_tensor_model_parallel_world_size,
     tensor_model_parallel_all_gather,
 )
+from vllm.model_executor.layers.attention import Attention
 from vllm.model_executor.layers.fused_moe import FusedMoE
 from vllm.model_executor.layers.fused_moe.config import FusedMoEParallelConfig
 from vllm.model_executor.layers.layernorm import RMSNorm

vllm/model_executor/models/granite.py

@@ -31,11 +31,11 @@ import torch
 from torch import nn
 from transformers import GraniteConfig
 
-from vllm.attention.layer import Attention
 from vllm.compilation.decorators import support_torch_compile
 from vllm.config import CacheConfig, VllmConfig
 from vllm.distributed import get_pp_group, get_tensor_model_parallel_world_size
 from vllm.model_executor.layers.activation import SiluAndMul
+from vllm.model_executor.layers.attention import Attention
 from vllm.model_executor.layers.layernorm import RMSNorm
 from vllm.model_executor.layers.linear import (
     MergedColumnParallelLinear,

vllm/model_executor/models/granitemoe.py

@@ -31,7 +31,6 @@ from typing import Any
 import torch
 from torch import nn
 
-from vllm.attention.layer import Attention
 from vllm.compilation.decorators import support_torch_compile
 from vllm.config import CacheConfig, VllmConfig
 from vllm.distributed import (
@@ -39,6 +38,7 @@ from vllm.distributed import (
     get_tensor_model_parallel_world_size,
     tensor_model_parallel_all_gather,
 )
+from vllm.model_executor.layers.attention import Attention
 from vllm.model_executor.layers.fused_moe import FusedMoE
 from vllm.model_executor.layers.layernorm import RMSNorm
 from vllm.model_executor.layers.linear import (

vllm/model_executor/models/granitemoehybrid.py

@@ -9,11 +9,11 @@ import torch
 from torch import nn
 from transformers import GraniteMoeHybridConfig
 
-from vllm.attention.layer import Attention
 from vllm.compilation.decorators import support_torch_compile
 from vllm.config import CacheConfig, ModelConfig, VllmConfig
 from vllm.distributed import get_tensor_model_parallel_world_size
 from vllm.distributed.parallel_state import get_pp_group
+from vllm.model_executor.layers.attention import Attention
 from vllm.model_executor.layers.layernorm import RMSNorm
 from vllm.model_executor.layers.linear import QKVParallelLinear, RowParallelLinear
 from vllm.model_executor.layers.logits_processor import LogitsProcessor