[Kernel][Core] Add AWQ support to the Marlin kernel (#6612)

vllm/model_executor/layers/quantization/__init__.py

@@ -2,6 +2,7 @@ from typing import Dict, Type
 
 from vllm.model_executor.layers.quantization.aqlm import AQLMConfig
 from vllm.model_executor.layers.quantization.awq import AWQConfig
+from vllm.model_executor.layers.quantization.awq_marlin import AWQMarlinConfig
 from vllm.model_executor.layers.quantization.base_config import (
     QuantizationConfig)
 from vllm.model_executor.layers.quantization.bitsandbytes import (
@@ -31,6 +32,7 @@ QUANTIZATION_METHODS: Dict[str, Type[QuantizationConfig]] = {
     "marlin": MarlinConfig,
     "gptq_marlin_24": GPTQMarlin24Config,
     "gptq_marlin": GPTQMarlinConfig,
+    "awq_marlin": AWQMarlinConfig,
     "gptq": GPTQConfig,
     "squeezellm": SqueezeLLMConfig,
     "compressed-tensors": CompressedTensorsConfig,