[Core] Support tensor parallelism for GGUF quantization (#7520)

vllm/model_executor/layers/quantization/gguf.py

@@ -5,7 +5,6 @@ import torch
 from torch.nn.parameter import Parameter, UninitializedParameter
 
 from vllm import _custom_ops as ops
-from vllm.distributed import get_tensor_model_parallel_world_size
 from vllm.model_executor.layers.linear import LinearBase, LinearMethodBase
 from vllm.model_executor.layers.quantization.base_config import (
     QuantizationConfig, QuantizeMethodBase)
@@ -39,9 +38,6 @@ class GGUFConfig(QuantizationConfig):
 
     @classmethod
     def from_config(cls, config: Dict[str, Any]) -> "GGUFConfig":
-        if get_tensor_model_parallel_world_size() > 1:
-            raise ValueError(
-                "GGUF quantization hasn't supported tensor parallelism yet.")
         return cls()
 
     def get_quant_method(self, layer: torch.nn.Module,