[6/N] pass whole config to inner model (#10205)

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

vllm/model_executor/models/arctic.py

@@ -34,7 +34,8 @@ from vllm.transformers_utils.configs.arctic import ArcticConfig
 
 from .interfaces import SupportsPP
 from .utils import (is_pp_missing_parameter,
-                    make_empty_intermediate_tensors_factory, make_layers)
+                    make_empty_intermediate_tensors_factory, make_layers,
+                    maybe_prefix)
 
 logger = init_logger(__name__)
 
@@ -364,14 +365,13 @@ class ArcticDecoderLayer(nn.Module):
 @support_torch_compile
 class ArcticModel(nn.Module):
 
-    def __init__(
-        self,
-        config: ArcticConfig,
-        cache_config: Optional[CacheConfig] = None,
-        quant_config: Optional[QuantizationConfig] = None,
-        prefix: str = "",
-    ) -> None:
+    def __init__(self, *, vllm_config: VllmConfig, prefix: str = ""):
         super().__init__()
+
+        config = vllm_config.model_config.hf_config
+        cache_config = vllm_config.cache_config
+        quant_config = vllm_config.quant_config
+
         self.padding_idx = config.pad_token_id
         self.vocab_size = config.vocab_size
         self.embed_tokens = VocabParallelEmbedding(
@@ -418,13 +418,10 @@ class ArcticForCausalLM(nn.Module, SupportsPP):
     def __init__(self, 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
         self.config = config
-        self.model = ArcticModel(config,
-                                 cache_config,
-                                 quant_config,
-                                 prefix=prefix)
+        self.model = ArcticModel(vllm_config=vllm_config,
+                                 prefix=maybe_prefix(prefix, "model"))
         self.vocab_size = config.vocab_size
         self.lm_head = ParallelLMHead(
             self.vocab_size,