[Quant] Arctic SupportsQuant (#13366)

vllm/model_executor/models/arctic.py

@@ -33,7 +33,7 @@ from vllm.model_executor.utils import set_weight_attrs
 from vllm.sequence import IntermediateTensors
 from vllm.transformers_utils.configs.arctic import ArcticConfig
 
-from .interfaces import SupportsPP
+from .interfaces import SupportsPP, SupportsQuant
 from .utils import (extract_layer_index, is_pp_missing_parameter,
                     make_empty_intermediate_tensors_factory, make_layers,
                     maybe_prefix)
@@ -423,7 +423,8 @@ class ArcticModel(nn.Module):
         return hidden_states
 
 
-class ArcticForCausalLM(nn.Module, SupportsPP):
+class ArcticForCausalLM(nn.Module, SupportsPP, SupportsQuant):
+    packed_modules_mapping = {"qkv_proj": ["q_proj", "k_proj", "v_proj"]}
 
     def __init__(self, *, vllm_config: VllmConfig, prefix: str = ""):
         super().__init__()