From df5407fb53b80a680b8cfa0d7b8646ca9ae80d40 Mon Sep 17 00:00:00 2001 From: Rain Jiang <96632942+rainj-me@users.noreply.github.com> Date: Mon, 8 Sep 2025 18:11:15 -0700 Subject: [PATCH] Revert "feat: add fused moe config for Qwen3-30B-A3B on B200" (#10185) --- ...dtype=fp8_w8a8,block_shape=[128, 128].json | 146 ------------------ 1 file changed, 146 deletions(-) delete mode 100644 python/sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=128,N=768,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json diff --git a/python/sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=128,N=768,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json b/python/sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=128,N=768,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json deleted file mode 100644 index b9dc2d71f..000000000 --- a/python/sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=128,N=768,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json +++ /dev/null @@ -1,146 +0,0 @@ -{ - "1": { - "BLOCK_SIZE_M": 16, - "BLOCK_SIZE_N": 128, - "BLOCK_SIZE_K": 128, - "GROUP_SIZE_M": 1, - "num_warps": 4, - "num_stages": 4 - }, - "2": { - "BLOCK_SIZE_M": 16, - "BLOCK_SIZE_N": 256, - "BLOCK_SIZE_K": 128, - "GROUP_SIZE_M": 1, - "num_warps": 4, - "num_stages": 5 - }, - "4": { - "BLOCK_SIZE_M": 16, - "BLOCK_SIZE_N": 128, - "BLOCK_SIZE_K": 256, - "GROUP_SIZE_M": 1, - "num_warps": 4, - "num_stages": 4 - }, - "8": { - "BLOCK_SIZE_M": 16, - "BLOCK_SIZE_N": 128, - "BLOCK_SIZE_K": 128, - "GROUP_SIZE_M": 1, - "num_warps": 4, - "num_stages": 4 - }, - "16": { - "BLOCK_SIZE_M": 16, - "BLOCK_SIZE_N": 128, - "BLOCK_SIZE_K": 256, - "GROUP_SIZE_M": 1, - "num_warps": 4, - "num_stages": 4 - }, - "24": { - "BLOCK_SIZE_M": 16, - "BLOCK_SIZE_N": 128, - "BLOCK_SIZE_K": 256, - "GROUP_SIZE_M": 1, - "num_warps": 4, - "num_stages": 4 - }, - "32": { - "BLOCK_SIZE_M": 16, - "BLOCK_SIZE_N": 128, - "BLOCK_SIZE_K": 256, - "GROUP_SIZE_M": 1, - "num_warps": 4, - "num_stages": 4 - }, - "48": { - "BLOCK_SIZE_M": 16, - "BLOCK_SIZE_N": 128, - "BLOCK_SIZE_K": 128, - "GROUP_SIZE_M": 1, - "num_warps": 4, - "num_stages": 4 - }, - "64": { - "BLOCK_SIZE_M": 16, - "BLOCK_SIZE_N": 128, - "BLOCK_SIZE_K": 256, - "GROUP_SIZE_M": 1, - "num_warps": 4, - "num_stages": 4 - }, - "96": { - "BLOCK_SIZE_M": 16, - "BLOCK_SIZE_N": 128, - "BLOCK_SIZE_K": 256, - "GROUP_SIZE_M": 1, - "num_warps": 4, - "num_stages": 4 - }, - "128": { - "BLOCK_SIZE_M": 16, - "BLOCK_SIZE_N": 128, - "BLOCK_SIZE_K": 256, - "GROUP_SIZE_M": 1, - "num_warps": 4, - "num_stages": 4 - }, - "256": { - "BLOCK_SIZE_M": 16, - "BLOCK_SIZE_N": 128, - "BLOCK_SIZE_K": 128, - "GROUP_SIZE_M": 1, - "num_warps": 4, - "num_stages": 3 - }, - "512": { - "BLOCK_SIZE_M": 64, - "BLOCK_SIZE_N": 128, - "BLOCK_SIZE_K": 256, - "GROUP_SIZE_M": 32, - "num_warps": 4, - "num_stages": 4 - }, - "1024": { - "BLOCK_SIZE_M": 64, - "BLOCK_SIZE_N": 128, - "BLOCK_SIZE_K": 128, - "GROUP_SIZE_M": 32, - "num_warps": 4, - "num_stages": 4 - }, - "1536": { - "BLOCK_SIZE_M": 64, - "BLOCK_SIZE_N": 128, - "BLOCK_SIZE_K": 128, - "GROUP_SIZE_M": 1, - "num_warps": 4, - "num_stages": 4 - }, - "2048": { - "BLOCK_SIZE_M": 64, - "BLOCK_SIZE_N": 128, - "BLOCK_SIZE_K": 256, - "GROUP_SIZE_M": 32, - "num_warps": 4, - "num_stages": 4 - }, - "3072": { - "BLOCK_SIZE_M": 64, - "BLOCK_SIZE_N": 128, - "BLOCK_SIZE_K": 128, - "GROUP_SIZE_M": 32, - "num_warps": 4, - "num_stages": 4 - }, - "4096": { - "BLOCK_SIZE_M": 64, - "BLOCK_SIZE_N": 128, - "BLOCK_SIZE_K": 256, - "GROUP_SIZE_M": 16, - "num_warps": 4, - "num_stages": 4 - } -}