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

Fix linear.py and improve weight loading (#2851)

Browse Source 
Co-authored-by: SangBin Cho <rkooo567@gmail.com>
This commit is contained in:
Lianmin Zheng
2025-01-13 01:39:14 -08:00
committed by GitHub
parent 4093aa4660
commit 72c7776355
12 changed files with 113 additions and 125 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
python/sglang/srt/layers/quantization/fp8_utils.py
View File

@@ -1,8 +1,8 @@
from typing import List, Optional, Tuple
import torch
from vllm.model_executor.parameter import RowvLLMParameter, _ColumnvLLMParameter
from sglang.srt.layers.parameter import RowvLLMParameter, _ColumnvLLMParameter
from sglang.srt.layers.quantization.fp8_kernel import (
per_token_group_quant_fp8,
w8a8_block_fp8_matmul,

2
python/sglang/srt/layers/quantization/modelopt_quant.py
View File

@@ -11,9 +11,9 @@ from vllm.model_executor.layers.quantization.utils.w8a8_utils import (
cutlass_fp8_supported,
requantize_with_max_scale,
)
from vllm.model_executor.parameter import ModelWeightParameter, PerTensorScaleParameter
from sglang.srt.layers.linear import LinearMethodBase
from sglang.srt.layers.parameter import ModelWeightParameter, PerTensorScaleParameter
from sglang.srt.layers.quantization.base_config import (
QuantizationConfig,
QuantizeMethodBase,