[5/N] pass the whole config to model (#9983)

Signed-off-by: youkaichao <youkaichao@gmail.com>

vllm/model_executor/models/opt.py

@@ -24,7 +24,7 @@ from transformers import OPTConfig
 
 from vllm.attention import Attention, AttentionMetadata
 from vllm.compilation.decorators import support_torch_compile
-from vllm.config import CacheConfig
+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.linear import (ColumnParallelLinear,
@@ -344,11 +344,13 @@ class OPTForCausalLM(nn.Module, SupportsPP):
 
     def __init__(
         self,
-        config: OPTConfig,
-        cache_config: Optional[CacheConfig] = None,
-        quant_config: Optional[QuantizationConfig] = None,
+        vllm_config: VllmConfig,
         prefix: str = "",
-    ):
+    ) -> None:
+        super().__init__()
+        config = vllm_config.model_config.hf_config
+        cache_config = vllm_config.cache_config
+        quant_config = vllm_config.quant_config
         super().__init__()
         self.config = config
         self.quant_config = quant_config