EngineX-Hygon/sglang
5
0
Fork 0
Code Issues Pull Requests Actions 7 Projects Releases Wiki Activity

[code style] Clean dead triton kernel code in fused_moe and useless vllm_ops import (#8310)

Browse Source
This commit is contained in:
Xiaoyu Zhang
2025-07-24 14:38:30 +08:00
committed by GitHub
parent 2f86f3ad62
commit a167fd0bcb
3 changed files with 27 additions and 242 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
python/sglang/srt/layers/quantization/utils.py
View File

@@ -17,15 +17,6 @@ from sglang.srt.utils import cpu_has_amx_support, is_cpu, is_cuda, is_hip, is_np
if TYPE_CHECKING:
from sglang.srt.layers.quantization.base_config import QuantizationConfig
_is_cuda = is_cuda()
_is_npu = is_npu()
_is_cpu_amx_available = cpu_has_amx_support()
_is_cpu = is_cpu()
_is_hip = is_hip()
if not (_is_cuda or _is_npu or (_is_cpu and _is_cpu_amx_available) or _is_hip):
from vllm._custom_ops import scaled_fp8_quant
def is_layer_skipped(
prefix: str,