Remove unused kwargs from model definitions (#13555)

vllm/model_executor/models/blip2.py

@@ -1,7 +1,7 @@
 # SPDX-License-Identifier: Apache-2.0
 
 from functools import cached_property
-from typing import (Iterable, List, Literal, Mapping, Optional, Set, Tuple,
+from typing import (Iterable, Literal, Mapping, Optional, Set, Tuple,
                     TypedDict, Union)
 
 import torch
@@ -9,7 +9,6 @@ import torch.nn as nn
 from transformers import (BatchFeature, Blip2Config, Blip2QFormerConfig,
                           apply_chunking_to_forward)
 
-from vllm.attention import AttentionMetadata
 from vllm.config import CacheConfig, VllmConfig
 from vllm.model_executor.layers.activation import get_act_fn
 from vllm.model_executor.layers.quantization import QuantizationConfig
@@ -658,8 +657,6 @@ class Blip2ForConditionalGeneration(nn.Module, SupportsMultiModal, SupportsPP):
         self,
         input_ids: torch.Tensor,
         positions: torch.Tensor,
-        kv_caches: List[torch.Tensor],
-        attn_metadata: AttentionMetadata,
         intermediate_tensors: Optional[IntermediateTensors] = None,
         inputs_embeds: Optional[torch.Tensor] = None,
         **kwargs: object,
@@ -708,8 +705,6 @@ class Blip2ForConditionalGeneration(nn.Module, SupportsMultiModal, SupportsPP):
 
         hidden_states = self.language_model.model(input_ids,
                                                   positions,
-                                                  kv_caches,
-                                                  attn_metadata,
                                                   intermediate_tensors,
                                                   inputs_embeds=inputs_embeds)