A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 8, 17:45 UTC
[Bugfix] Fix _fused_moe_lora_expand signature mismatch (#33821)
Signed-off-by: Xin Yang <xyangx@amazon.com>
Xin Yang committed
Feb 6, 2026 at 18:45 UTC
3920cafdd6a8b20a76fa4805e010ae5da8f3a45f
1 file changed
-1
vllm/lora/ops/triton_ops/fused_moe_lora_op.py
-1
@@ -779,7 +779,6 @@ def _fused_moe_lora_shrink_fake(
779
def _fused_moe_lora_expand_fake(
780
output: torch.Tensor,
781
a_intermediate_cache1: torch.Tensor,
782
- b_intermediate_cache1: torch.Tensor,
782
lora_b_stacked: list[torch.Tensor],
783
topk_weights: torch.Tensor,
784
sorted_token_ids: torch.Tensor | None,