Commit 7c51c698 authored by zhuwenwen's avatar zhuwenwen
Browse files

update List[int]

parent 0798f59e
...@@ -1120,7 +1120,7 @@ def flashinfer_fused_moe_blockscale_fp8( ...@@ -1120,7 +1120,7 @@ def flashinfer_fused_moe_blockscale_fp8(
intermediate_size: int, intermediate_size: int,
expert_offset: int, expert_offset: int,
local_num_experts: int, local_num_experts: int,
block_shape: list[int], block_shape: List[int],
routed_scaling: float = 1.0) -> torch.Tensor: routed_scaling: float = 1.0) -> torch.Tensor:
from vllm.utils.flashinfer import flashinfer_trtllm_fp8_block_scale_moe from vllm.utils.flashinfer import flashinfer_trtllm_fp8_block_scale_moe
assert top_k <= global_num_experts assert top_k <= global_num_experts
...@@ -1173,7 +1173,7 @@ def flashinfer_fused_moe_blockscale_fp8_fake( ...@@ -1173,7 +1173,7 @@ def flashinfer_fused_moe_blockscale_fp8_fake(
intermediate_size: int, intermediate_size: int,
expert_offset: int, expert_offset: int,
local_num_experts: int, local_num_experts: int,
block_shape: list[int], block_shape: List[int],
routed_scaling: float = 1.0) -> torch.Tensor: routed_scaling: float = 1.0) -> torch.Tensor:
return torch.empty_like(x) return torch.empty_like(x)
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment