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

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

vllm/model_executor/models/baichuan.py

@@ -29,7 +29,6 @@ import torch
 from torch import nn
 from transformers import PretrainedConfig
 
-from vllm.attention.layer import Attention
 from vllm.compilation.decorators import support_torch_compile
 from vllm.config import CacheConfig, VllmConfig
 from vllm.distributed import (
@@ -38,6 +37,7 @@ from vllm.distributed import (
     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/bailing_moe.py

@@ -32,7 +32,6 @@ import torch.nn.functional as F
 from torch import nn
 from transformers.configuration_utils import PretrainedConfig
 
-from vllm.attention.layer import Attention
 from vllm.compilation.decorators import support_torch_compile
 from vllm.config import CacheConfig, VllmConfig
 from vllm.distributed import (
@@ -41,6 +40,7 @@ from vllm.distributed import (
     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.fused_moe import SharedFusedMoE
 from vllm.model_executor.layers.layernorm import RMSNorm
 from vllm.model_executor.layers.linear import (

vllm/model_executor/models/bamba.py

@@ -9,12 +9,12 @@ import torch
 from torch import nn
 from transformers import BambaConfig
 
-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/bert.py

@@ -11,7 +11,7 @@ from vllm.compilation.decorators import support_torch_compile
 from vllm.config import CacheConfig, ModelConfig, PoolerConfig, VllmConfig
 from vllm.distributed import get_tensor_model_parallel_world_size
 from vllm.model_executor.layers.activation import get_act_fn
-from vllm.model_executor.layers.attention.encoder_only_attention import (
+from vllm.model_executor.layers.attention import (
     EncoderOnlyAttention,
 )
 from vllm.model_executor.layers.linear import (

vllm/model_executor/models/bert_with_rope.py

@@ -15,7 +15,7 @@ from vllm.distributed import (
     tensor_model_parallel_all_reduce,
 )
 from vllm.model_executor.layers.activation import get_act_and_mul_fn, get_act_fn
-from vllm.model_executor.layers.attention.encoder_only_attention import (
+from vllm.model_executor.layers.attention import (
     EncoderOnlyAttention,
 )
 from vllm.model_executor.layers.fused_moe import activation_without_mul, fused_topk

vllm/model_executor/models/blip.py

@@ -11,7 +11,7 @@ from transformers import Blip2VisionConfig, BlipVisionConfig
 
 from vllm.distributed import divide, get_tensor_model_parallel_world_size
 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.conv import Conv2dLayer
 from vllm.model_executor.layers.linear import (
     ColumnParallelLinear,

vllm/model_executor/models/bloom.py

@@ -27,7 +27,6 @@ import torch
 from torch import nn
 from transformers import BloomConfig
 
-from vllm.attention.layer import Attention
 from vllm.compilation.decorators import support_torch_compile
 from vllm.config import CacheConfig, VllmConfig
 from vllm.distributed import (
@@ -36,6 +35,7 @@ from vllm.distributed 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/chameleon.py

@@ -16,12 +16,12 @@ from transformers import (
     ChameleonVQVAEConfig,
 )
 
-from vllm.attention.layer import Attention
 from vllm.config import CacheConfig, VllmConfig
 from vllm.config.multimodal import BaseDummyOptions
 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 SiluAndMul
+from vllm.model_executor.layers.attention import Attention
 from vllm.model_executor.layers.conv import Conv2dLayer
 from vllm.model_executor.layers.layernorm import RMSNorm
 from vllm.model_executor.layers.linear import (

vllm/model_executor/models/chatglm.py

@@ -12,11 +12,11 @@ import torch
 from torch import nn
 from torch.nn import LayerNorm
 
-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/clip.py

@@ -14,12 +14,11 @@ from transformers import (
     CLIPVisionConfig,
 )
 
-from vllm.attention.layer import Attention
 from vllm.config import VllmConfig
 from vllm.config.multimodal import BaseDummyOptions
 from vllm.distributed import divide, get_tensor_model_parallel_world_size
 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 Attention, MMEncoderAttention
 from vllm.model_executor.layers.conv import Conv2dLayer
 from vllm.model_executor.layers.linear import (
     ColumnParallelLinear,

vllm/model_executor/models/commandr.py

@@ -30,11 +30,11 @@ import torch
 from torch import nn
 from transformers import Cohere2Config, CohereConfig
 
-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.linear import (
     MergedColumnParallelLinear,
     QKVParallelLinear,

vllm/model_executor/models/dbrx.py

@@ -8,13 +8,13 @@ import torch
 import torch.nn as nn
 from transformers import DbrxConfig
 
-from vllm.attention.layer import Attention
 from vllm.config import CacheConfig, VllmConfig
 from vllm.distributed import (
     get_pp_group,
     get_tensor_model_parallel_rank,
     get_tensor_model_parallel_world_size,
 )
+from vllm.model_executor.layers.attention import Attention
 from vllm.model_executor.layers.fused_moe import FusedMoE
 from vllm.model_executor.layers.linear import (
     QKVParallelLinear,

vllm/model_executor/models/deepencoder.py

@@ -18,7 +18,7 @@ import torch.nn as nn
 import torch.nn.functional as F
 from transformers import CLIPVisionConfig
 
-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.quantization import QuantizationConfig
 from vllm.model_executor.model_loader.weight_utils import default_weight_loader

vllm/model_executor/models/deepseek_v2.py

@@ -33,7 +33,6 @@ from torch import nn
 from transformers import DeepseekV2Config, DeepseekV3Config
 
 from vllm._aiter_ops import rocm_aiter_ops
-from vllm.attention.layer import Attention
 from vllm.compilation.decorators import support_torch_compile
 from vllm.config import CacheConfig, ParallelConfig, VllmConfig, get_current_vllm_config
 from vllm.distributed import (
@@ -45,6 +44,7 @@ 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.attention_layer_base import AttentionLayerBase
 from vllm.model_executor.layers.fused_moe import SharedFusedMoE
 from vllm.model_executor.layers.layernorm import LayerNorm, RMSNorm

vllm/model_executor/models/dots1.py

@@ -32,7 +32,6 @@ import torch
 from torch import nn
 from transformers import Dots1Config
 
-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 (
@@ -41,6 +40,7 @@ from vllm.distributed import (
     tensor_model_parallel_all_reduce,
 )
 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/dots_ocr.py

@@ -16,7 +16,7 @@ from vllm.distributed.parallel_state import (
     get_tensor_model_parallel_world_size,
 )
 from vllm.model_executor.layers.activation import SiluAndMul
-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

vllm/model_executor/models/ernie45_moe.py

@@ -32,7 +32,6 @@ import torch
 from torch import nn
 from transformers import PretrainedConfig
 
-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/ernie45_vl.py

@@ -42,7 +42,7 @@ from vllm.distributed import parallel_state
 from vllm.distributed import utils as dist_utils
 from vllm.logger import init_logger
 from vllm.model_executor.layers.activation import QuickGELU
-from vllm.model_executor.layers.attention.mm_encoder_attention import (
+from vllm.model_executor.layers.attention import (
     MMEncoderAttention,
 )
 from vllm.model_executor.layers.layernorm import RMSNorm

vllm/model_executor/models/ernie45_vl_moe.py

@@ -31,12 +31,11 @@ import torch
 from torch import nn
 from transformers import PretrainedConfig
 
-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.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/exaone.py

@@ -32,11 +32,11 @@ import torch
 from torch import nn
 from transformers import PretrainedConfig
 
-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,