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

chore: cleanup quant deps (#12268)

Browse Source
This commit is contained in:
Yineng Zhang
2025-10-28 02:03:57 -07:00
committed by GitHub
parent ea39952797
commit 64cf868eba
2 changed files with 7 additions and 80 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
python/sglang/srt/model_executor/model_runner.py
View File

@@ -77,7 +77,6 @@ from sglang.srt.layers.dp_attention import (
initialize_dp_attention,
)
from sglang.srt.layers.logits_processor import LogitsProcessorOutput
from sglang.srt.layers.quantization import monkey_patch_isinstance_for_vllm_base_layer
from sglang.srt.layers.sampler import Sampler
from sglang.srt.layers.torchao_utils import apply_torchao_config_to_model
from sglang.srt.lora.lora_manager import LoRAManager
@@ -730,7 +729,6 @@ class ModelRunner:
# Load the model
# Remove monkey_patch when linear.py quant remove dependencies with vllm
monkey_patch_vllm_parallel_state()
monkey_patch_isinstance_for_vllm_base_layer()
with self.memory_saver_adapter.region(
GPU_MEMORY_TYPE_WEIGHTS,
@@ -742,7 +740,6 @@ class ModelRunner:
device_config=DeviceConfig(self.device, self.gpu_id),
)
monkey_patch_vllm_parallel_state(reverse=True)
monkey_patch_isinstance_for_vllm_base_layer(reverse=True)
get_offloader().post_init()