minke.yu/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 12 Packages Projects Releases Wiki Activity

Fix gpt-oss-20b with mxfp4 support for Xeon (#27782)

Browse Source
This commit is contained in:
YanbingJiang
2026-06-11 09:16:16 +08:00
committed by GitHub
parent e113d884d0
commit 66989a7642
1 changed files with 3 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
python/sglang/srt/layers/quantization/__init__.py
+3 -1
View File
@@ -55,6 +55,7 @@ from sglang.srt.layers.quantization.w8a8_int8 import W8A8Int8Config
from sglang.srt.platforms import current_platform from sglang.srt.platforms import current_platform
from sglang.srt.utils import ( from sglang.srt.utils import (
cpu_has_amx_support, cpu_has_amx_support,
is_cpu,
is_cuda, is_cuda,
is_hip, is_hip,
is_mps, is_mps,
@@ -99,7 +100,7 @@ BASE_QUANTIZATION_METHODS: Dict[str, Type[QuantizationConfig]] = {
} }
if is_cuda() or (_is_mxfp_supported and is_hip()): if is_cpu() or is_cuda() or (_is_mxfp_supported and is_hip()):
BASE_QUANTIZATION_METHODS.update( BASE_QUANTIZATION_METHODS.update(
{ {
"mxfp4": Mxfp4Config, "mxfp4": Mxfp4Config,
@@ -130,6 +131,7 @@ CPU_QUANTIZATION_METHODS = {
"compressed-tensors": CompressedTensorsConfig, "compressed-tensors": CompressedTensorsConfig,
"awq": AWQCPUConfig, "awq": AWQCPUConfig,
"gptq": CPUGPTQConfig, "gptq": CPUGPTQConfig,
"mxfp4": Mxfp4Config,
} }
QUANTIZATION_METHODS = {**BASE_QUANTIZATION_METHODS} QUANTIZATION_METHODS = {**BASE_QUANTIZATION_METHODS}