diff --git a/flashinfer/cascade.py b/flashinfer/cascade.py index 38458bb9..20b91259 100644 --- a/flashinfer/cascade.py +++ b/flashinfer/cascade.py @@ -317,10 +317,10 @@ def __init__( float_workspace_buffer: torch.Tensor, kv_layout: str = "NHD", use_cuda_graph: bool = False, - qo_indptr_buf_arr: Optional[list[torch.Tensor]] = None, - paged_kv_indptr_buf_arr: Optional[list[torch.Tensor]] = None, - paged_kv_indices_buf_arr: Optional[list[torch.Tensor]] = None, - paged_kv_last_page_len_buf_arr: Optional[list[torch.Tensor]] = None, + qo_indptr_buf_arr: Optional[List[torch.Tensor]] = None, + paged_kv_indptr_buf_arr: Optional[List[torch.Tensor]] = None, + paged_kv_indices_buf_arr: Optional[List[torch.Tensor]] = None, + paged_kv_last_page_len_buf_arr: Optional[List[torch.Tensor]] = None, ) -> None: r"""Constructor of :class:`MultiLevelCascadeAttentionWrapper`.