Remove unused kwargs from model definitions (#13555)

vllm/model_executor/models/phi3v.py

@@ -23,7 +23,6 @@ import torch.nn as nn
 from transformers import (BatchFeature, CLIPVisionConfig, PretrainedConfig,
                           ProcessorMixin)
 
-from vllm.attention import AttentionMetadata
 from vllm.config import VllmConfig
 from vllm.logger import init_logger
 from vllm.model_executor.layers.quantization import QuantizationConfig
@@ -672,8 +671,6 @@ class Phi3VForCausalLM(nn.Module, SupportsMultiModal, SupportsPP,
     def forward(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):
@@ -691,8 +688,6 @@ class Phi3VForCausalLM(nn.Module, SupportsMultiModal, SupportsPP,
 
         hidden_states = self.language_model.model(input_ids,
                                                   positions,
-                                                  kv_caches,
-                                                  attn_metadata,
                                                   intermediate_tensors,
                                                   inputs_embeds=inputs_embeds)