[Kernel] Migrate scattered quantization kernels to sglang.kernels (RFC #29630, Phase 2.5, 1/7) (#30784)

Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
2026-07-13 16:17:01 +08:00
parent 2225817424
commit 874fc07d9b
252 changed files with 294 additions and 214 deletions
@@ -3,10 +3,10 @@ import unittest

 import torch

+from sglang.kernels.ops.quantization.int8_kernel import per_token_quant_int8
 from sglang.srt.layers.activation import SiluAndMul
 from sglang.srt.layers.moe.moe_runner.triton_utils.fused_moe import fused_moe
 from sglang.srt.layers.moe.topk import TopKConfig, select_experts
-from sglang.srt.layers.quantization.int8_kernel import per_token_quant_int8
 from sglang.srt.server_args import ServerArgs, set_global_server_args_for_scheduler
 from sglang.test.ci.ci_register import register_amd_ci, register_cuda_ci
 from sglang.test.test_utils import CustomTestCase