[1/2] sgl-kernel: Fuse routed scaling factor into select_experts (#8364)

2025-08-01 18:14:24 -07:00
parent 82e6c3a65a
commit f642524fd9
5 changed files with 38 additions and 12 deletions
--- a/sgl-kernel/include/sgl_kernel_ops.h
+++ b/sgl-kernel/include/sgl_kernel_ops.h
@@ -243,7 +243,8 @@ std::vector<at::Tensor> moe_fused_gate(
    int64_t topk_group,
    int64_t topk,
    int64_t num_fused_shared_experts,
-    double routed_scaling_factor);
+    double routed_scaling_factor,
+    bool apply_routed_scaling_factor_on_output);

 void fp8_blockwise_scaled_grouped_mm(
    torch::Tensor& output,