[XPU] Route grouped_topk to the fused _moe_C kernel on XPU (#53580)

Signed-off-by: Marceli Fylcek <marceli.fylcek@intel.com>
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
This commit is contained in:
Marceli Fylcek
2026-09-07 17:43:38 +08:00
committed by GitHub
co-authored by Claude Opus 5
parent 7dbe386833
commit 9b85112e13
2 changed files with 23 additions and 3 deletions
@@ -90,7 +90,7 @@ def grouped_topk(
) -> tuple[torch.Tensor, torch.Tensor]:
if (
envs.VLLM_USE_FUSED_MOE_GROUPED_TOPK
and current_platform.is_cuda()
and (current_platform.is_cuda() or current_platform.is_xpu())
and num_expert_group <= 32
and topk <= 32
and e_score_correction_bias is not None