fix optional and list

vllm/config/model.py

@@ -297,7 +297,7 @@ class ModelConfig:
     definitions"""
     io_processor_plugin: str | None = None
     """IOProcessor plugin name to load at model startup"""
-    enable_chunked_prefill: Optional[bool] = None
+    enable_chunked_prefill: bool | None = None
     """If True, prefill requests can be chunked based
     on the remaining max_num_batched_tokens."""
 

vllm/model_executor/layers/quantization/utils/fp8_utils.py

@@ -54,7 +54,7 @@ def cutlass_scaled_mm(
     B: torch.Tensor,
     As: torch.Tensor,
     Bs: torch.Tensor,
-    block_size: List[int],
+    block_size: list[int],
     output_dtype: torch.dtype = torch.float16,
 ) -> torch.Tensor:
     return ops.cutlass_scaled_mm(
@@ -787,7 +787,7 @@ def w8a8_triton_block_scaled_mm(
     B: torch.Tensor,
     As: torch.Tensor,
     Bs: torch.Tensor,
-    block_size: List[int],
+    block_size: list[int],
     output_dtype: torch.dtype = torch.float16,
 ) -> torch.Tensor:
     """This function performs matrix multiplication with block-wise