From 97e86fd55922e3b510ac5014387c3bc2096da065 Mon Sep 17 00:00:00 2001 From: elvischenv <219235043+elvischenv@users.noreply.github.com> Date: Tue, 23 Dec 2025 17:52:53 -0800 Subject: [PATCH] remove tile_tokens_dim Signed-off-by: elvischenv <219235043+elvischenv@users.noreply.github.com> --- .../layers/quantization/utils/flashinfer_fp4_moe.py | 1 - 1 file changed, 1 deletion(-) diff --git a/vllm/model_executor/layers/quantization/utils/flashinfer_fp4_moe.py b/vllm/model_executor/layers/quantization/utils/flashinfer_fp4_moe.py index dceee42f31e39..4611b83757a69 100644 --- a/vllm/model_executor/layers/quantization/utils/flashinfer_fp4_moe.py +++ b/vllm/model_executor/layers/quantization/utils/flashinfer_fp4_moe.py @@ -325,7 +325,6 @@ def flashinfer_trtllm_fp4_moe( local_expert_offset=layer.ep_rank * layer.local_num_experts, local_num_experts=layer.local_num_experts, routed_scaling_factor=None, - tile_tokens_dim=None, routing_method_type=routing_method_type, do_finalize=True, )[0]